WO2020237495A1 - 地面可移动平台与其运动信息检测方法、系统 - Google Patents
地面可移动平台与其运动信息检测方法、系统 Download PDFInfo
- Publication number
- WO2020237495A1 WO2020237495A1 PCT/CN2019/088757 CN2019088757W WO2020237495A1 WO 2020237495 A1 WO2020237495 A1 WO 2020237495A1 CN 2019088757 W CN2019088757 W CN 2019088757W WO 2020237495 A1 WO2020237495 A1 WO 2020237495A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- movable platform
- data
- ground movable
- ground
- movement information
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C23/00—Combined instruments indicating more than one navigational value, e.g. for aircraft; Combined measuring devices for measuring two or more variables of movement, e.g. distance, speed or acceleration
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
- G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
- G01C21/16—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
- G01C21/165—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation combined with non-inertial navigation instruments
Definitions
- This application relates to the field of intelligent driving technology, and in particular to a ground movable platform and its motion information detection method and system.
- the vehicle's own position and posture recognition is one of the key technologies of automatic driving, and its purpose is to use the sensors on the car to recognize the vehicle's own position and posture.
- the accuracy and reliability of the vehicle's own position and attitude recognition results directly determine the accuracy and reliability of subsequent sensor data processing, vehicle navigation and control. How to balance high-precision and high-reliability pose recognition has become a key technical issue in this field.
- the method based on visual synchronization positioning and map construction (SLAM, Simultaneous Localization And Mapping), visual inertial navigation system (VINS, Visual Inertial Navigation System) and related data fusion methods are commonly used as a method of pose recognition.
- the speed direction is parallel to the direction of the head of the vehicle indicated by the Inertial Measurement Unit (IMU), and on this basis, the image feature points in the image are detected and tracked to recover the three-dimensional motion information, thereby obtaining the vehicle's pose recognition result.
- IMU Inertial Measurement Unit
- the inertial sensors in the vehicle may cause the pose recognition result to have components outside the actual constrained direction due to attitude drift and visual environmental errors, which makes the recognition accuracy and reliability of the pose recognition result.
- the embodiment of the present invention provides a ground movable platform and its motion information detection method and system, which are used to improve the accuracy and reliability of the pose recognition result.
- an embodiment of the present invention provides a method for detecting movement information of a ground movable platform, including:
- the image data including environmental image data of the surrounding environment of the ground movable platform
- the moving speed data is acquired by a wheel speed collecting device mounted on the movable platform on the ground.
- an embodiment of the present invention provides a device for detecting movement information of a ground movable platform, including:
- the computer program is stored in the memory and is configured to be executed by the processor to implement the method according to the first aspect.
- an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon,
- the computer program is executed by the processor to implement the method according to the first aspect.
- an embodiment of the present invention provides a movement information detection system for a ground movable platform, including:
- An image acquisition device for acquiring at least two frames of the image data
- An inertial sensor for acquiring the inertial navigation data of the ground movable platform
- a wheel speed acquisition device for acquiring the moving speed data of the ground movable platform
- Movement information detection device for ground movable platform used for:
- the moving speed data is acquired by the wheel speed collecting device mounted on the movable platform on the ground.
- an embodiment of the present invention provides a ground movable platform, including:
- the power system is installed on the fuselage to provide operating power
- a device for detecting movement information of a ground movable platform configured to execute the method as described in the first aspect
- An image acquisition device for acquiring at least two frames of the image data
- An inertial sensor for acquiring the inertial navigation data of the ground movable platform
- the wheel speed acquisition device is used to acquire the moving speed data of the ground movable platform.
- the ground movable platform and its motion information detection method and system provided by the embodiments of the present invention detect the motion function information of the ground movable platform through image data, the inertial navigation data and the moving speed data. Because the moving speed data is restricted The horizontal direction data on the road surface does not have a vertical component. Therefore, this solution uses the movement speed data to restrict the movement information of the ground movable platform in the horizontal direction, avoiding the movement information from the actual constraint direction due to the attitude drift of the IMU. The situation of the component of and the influence of this situation on the result of pose recognition, which is beneficial to improve the recognition accuracy and the reliability of the recognition result.
- Figure 1 is a schematic diagram of an uncalibrated tilt angle involved in an embodiment of the present invention
- FIG. 2 is a schematic flowchart of a method for detecting movement information of a ground movable platform according to an embodiment of the present invention
- FIG. 3 is a schematic flowchart of another method for detecting movement information of a ground movable platform according to an embodiment of the present invention
- FIG. 4 is a schematic flowchart of another method for detecting movement information of a ground movable platform according to an embodiment of the present invention
- FIG. 5 is a schematic flowchart of another method for detecting movement information of a ground movable platform according to an embodiment of the present invention
- FIG. 6 is a schematic flowchart of another method for detecting movement information of a ground movable platform according to an embodiment of the present invention.
- FIG. 7 is a functional block diagram of a device for detecting movement information of a ground movable platform provided by an embodiment of the present invention.
- FIG. 8 is a schematic diagram of the physical structure of a device for detecting movement information of a ground movable platform provided by an embodiment of the present invention.
- FIG. 9 is a schematic diagram of the architecture of a system for detecting movement information of a ground movable platform provided by an embodiment of the present invention.
- FIG. 10 is a schematic diagram of the architecture of a ground movable platform provided by an embodiment of the present invention.
- the specific application scenario of the present invention is the location and posture recognition scenario for the ground movable platform. Furthermore, it can be an automatic driving or automatic control system for a movable platform on the ground.
- the ground movable platform involved in the embodiment of the present invention may include, but is not limited to, a vehicle.
- other ground machinery devices may also be included, where the ground machinery devices may include, but are not limited to: at least one of ground machinery toys and ground robots.
- the ground machine toy may be a smart toy car, and for example, the ground robot may be a ground sweeping robot, a ground dispatching robot, and so on.
- the embodiment of the present invention takes a vehicle as an example for specific description.
- the existing method for implementing pose recognition based on the fusion of SLAM, VINS and related data is based on the assumption that the speed direction of the vehicle is parallel to the heading direction indicated by the IMU.
- the uncalibrated tilt angle error there is an error in the vertical direction between the vehicle's speed direction (wheel coordinate system) and the front direction (inertial coordinate system), that is, the uncalibrated tilt angle error.
- a suspension system is provided in the vehicle.
- the wheels and the vehicle body are not rigidly connected.
- the inclination angle between the car body and the road surface will be different.
- Figure 1 shows a schematic diagram of the uncalibrated pitch error of the vehicle. As shown in Figure 1, when the center of gravity of the vehicle is back, the front of the vehicle will be upward; when the center of gravity of the vehicle is forward, the front of the vehicle will be downward. At the same time, during driving, the inclination angle between the car body and the road will also oscillate.
- the technical solution provided by the present invention aims to solve the above technical problems of the prior art.
- the embodiment of the present invention provides a method for detecting movement information of a ground movable platform. Please refer to Figure 2.
- the method includes the following steps:
- S202 Acquire at least two frames of image data, where the image data includes environmental image data of the surrounding environment of the ground movable platform.
- the inertial navigation data can be obtained through an image acquisition device set on a movable platform on the ground.
- the image acquisition device may include, but is not limited to, a camera, where the camera may include, but is not limited to: a color camera, a grayscale camera, and the like.
- the embodiment of the present invention does not specifically limit the specific parameters of the image acquisition device.
- the number of cameras can be one or more, which can be a calibrated one or a pair of binocular cameras.
- the inertial navigation data can be measured by an inertial sensor (Inertial Measurement Unit, IMU) installed on a movable platform on the ground.
- IMU Inertial Measurement Unit
- the data may specifically include but is not limited to at least one of acceleration and angular velocity.
- the wheel speed collection device involved in the embodiment of the present invention may include but is not limited to at least one of the following: a wheel speed sensor and a wheel encoder. It should be understood that the wheel speed acquisition device involved in the embodiment of the present invention is not limited to the measurement of the movement speed of the wheel-driven ground movable platform, but also includes the measurement of the movement speed of the movable ground platform driven by other methods, such as crawler robots. .
- the movement speed data involved in the embodiment of the present invention may include, but is not limited to: wheel speed data or rotation angle data of at least one wheel of the ground movable platform. It can be seen that, before performing the step S208, the embodiment of the present invention may further include the following step: acquiring wheel speed data or turning angle data of at least one wheel collected by the wheel speed collecting device as the moving speed data.
- the aforementioned at least one wheel may include, but is not limited to: two rear wheels of the ground movable platform. That is, the wheel speed data of the two rear wheels (that is, the left rear wheel and the right rear wheel) of the ground movable platform can be acquired as the moving speed data.
- the embodiment of the present invention does not specifically limit which wheel the at least one wheel of the ground movable platform is.
- the wheel speed data or the turning angle data of the four wheels of the ground movable platform can also be obtained as the movement speed data; or, any of the ground movable platforms can also be obtained
- the rotation angle data of one front wheel is used as the moving speed data.
- the wheels can be preset according to actual needs, and will not be repeated.
- S208 Detect the movement information of the ground movable platform according to the image data, the inertial navigation data and the movement speed data of the ground movable platform.
- the position data, posture data and speed data of the ground movable platform in the world coordinate system are detected.
- the image data and the inertial navigation data may also be further considered to realize the detection of the movement information of the ground movable platform.
- the following describes the processing methods of the aforementioned three types of data respectively. It can be seen that the following description is implemented under the premise of the aforementioned inventive concept. Even if the motion information is detected by processing a certain type of data, it does not mean that other data is not considered. For example, the subsequent mention of “detecting the movement information of the ground movable platform based on the movement speed data” refers to the processing method of processing movement speed data to detect movement information, and does not mean that the data is solely relied on to achieve the The detection of the movement information of the ground movable platform.
- S208 can refer to the process shown in FIG. 3 when implementing S208.
- S208 includes the following steps:
- S2082 Use the environmental feature constraints of the ground movable platform to process the image data to detect movement information of the ground movable platform.
- the environmental feature constraints are used to restrict the environment where the ground movable platform is located.
- the corresponding relationship between the image data at different moments that is, the environment in the image data at similar moments has overlapping parts, therefore, the environment feature constraint can realize the constraint on the environment of the ground movable platform.
- it can implement environmental feature constraints through feature points between at least two frames of image data, where the feature points are feature points of the environment where the ground movable platform is located.
- S2084 Use the inertial feature constraint of the movable ground platform to process the inertial navigation data to detect movement information of the movable ground platform.
- the inertial feature constraint is used to constrain the position and attitude of the ground movable platform.
- the inertial navigation data collected by the inertial sensor can provide inertial feature constraints for the attitude of the ground movable platform.
- S2086 Use the plane feature constraints of the ground movable platform to process the movement speed data to detect the movement information of the ground movable platform.
- the plane feature constraint is used to constrain the actual pose of the ground movable platform in the horizontal direction.
- the movement speed data of the ground movable platform is the horizontal direction data constrained on the road surface, and there is no vertical component. Therefore, the embodiment of the present invention uses the movement speed data to perform the horizontal direction on the movement information of the ground movable platform.
- the above limitation avoids the vertical component of the motion information (that is, the component outside the actual constraint direction) caused by the posture drift of the IMU and the influence of this situation on the pose recognition result, thereby achieving improved recognition accuracy and recognition The effect of the reliability of the result.
- the aforementioned environmental feature constraints, inertial feature constraints, and plane feature constraints can be used as iterative solution conditions for nonlinear optimization problems. In this way, the nonlinear optimization problem can be solved to obtain the position, attitude and other motion information of the ground movable platform.
- S20822 Perform feature point matching on the at least two frames of image data to obtain environmental feature points.
- the feature points of each frame of image data are extracted, and then the feature points of each frame of image are matched. If the same feature points can be matched in at least two frames of images, these feature points can be used to characterize the movable ground The environmental characteristics of the environment where the platform is located. At this time, the matched characteristic points are used as environmental characteristic points.
- S20824 Use the environmental feature constraints to process the environmental feature points to detect the movement information of the ground movable platform.
- the aforementioned environmental feature points can be used to detect the movement information of the ground movable platform.
- step of using inertial feature constraints to process inertial navigation data in S2084, as shown in Figure 4, can be implemented in the following ways:
- S20842 Obtain speed data and/or displacement data of the ground movable platform according to the inertial navigation data.
- the inertial navigation data is processed to obtain speed data and/or displacement data of the ground movable platform.
- the processing method can include but is not limited to: integral algorithm.
- S20844 Use the inertial feature constraint of the movable ground platform to process the speed data and/or the displacement data to detect the movement information of the movable ground platform.
- the aforementioned speed data and/or displacement data can be used to detect the movement information of the movable platform on the ground.
- step of using plane feature constraints to process movement speed data in S2086 can be obtained by the method shown in FIG. 4.
- S2086 can be implemented as follows:
- S20862 Acquire pose data of the movable platform on the ground according to the movement speed data.
- the movement speed data is processed to obtain the pose data of the ground movable platform.
- the movement speed data between any two moments can be processed to obtain the pose data of the ground movable platform according to the movement speed between the two moments.
- the pose data of the ground movable platform is obtained by processing the movement speed data between the first moment (hereinafter referred to as k moment) and the second moment (hereinafter referred to as j moment).
- the obtained pose data of the ground movable platform can be: P_Wk_Wj and R_Wk_Wj; where P represents the position, R represents the attitude, W represents the wheel speed integral coordinate system, and P_Wk_Wj represents the ground movable platform
- P_Wk_Wj represents the ground movable platform
- the relative position of the wheel speed integral coordinate system between k and j, R_Wk_Wj represents the relative posture of the ground movable platform in the wheel speed integral coordinate system between k and j.
- S20864 Perform first coordinate conversion processing on the pose data to obtain inertial pose data in an inertial coordinate system.
- the pose data obtained by S20862 is the relative pose under the wheel speed integral coordinate system, and when specifically identifying the pose of the ground movable platform, what needs to be obtained is a certain moment (the present invention
- the embodiment assumes that it is time k) the position, posture, and velocity data of the inertial sensor in the world coordinate system.
- R_Ik_Ij represents the relative posture of the ground movable platform in the inertial coordinate system between time k and j
- P_Ik_Ij represents the relative position of the ground movable platform in the inertial coordinate system between time k and j.
- This coordinate conversion step can be implemented as shown in the following formula:
- R_Ik_Ij R_I_W ⁇ R_Wk_Wj ⁇ R_W_I
- P_Ik_Ij P_I_W+R_I_W ⁇ P_Wk_Wj–R_Ik_Ij ⁇ P_I_W
- R_I_W is used to represent the relative posture between the inertial coordinate system calculated by offline calibration and the wheel speed integral coordinate system
- R_W_I is the inverse matrix of R_I_W
- P_I_W is used to represent the inertial coordinate system and wheel speed integral coordinate system of offline calibration calculation
- the relative position between P_W_I is the inverse matrix of P_I_W.
- the horizontal component can be taken; and for the relative posture data obtained above, only the z-axis component (horizontal component) can be removed.
- the horizontal pose data in the inertial coordinate system obtained in this step is recorded as: R_Ik_Ij_H and P_Ik_Ij_H, where H represents the horizontal component, R_Ik_Ij_H is the horizontal component of R_Ik_Ij, and P_Ik_Ij_H is the horizontal component of P_Ik_Ij.
- the plane feature constraint includes a first relationship
- the first relationship is a relationship between the horizontal pose data and the actual pose.
- the first relationship can be characterized as the following formula:
- R_Ik_Ij_H R_G_Ik ⁇ (-1) ⁇ R_G_Ij ⁇ noise_R
- G represents the world coordinate system
- R_G_Ik represents the posture of the inertial sensor in the world coordinate system at time k (that is: the actual posture of the ground movable platform to be acquired)
- R_G_Ij represents the posture of the inertial sensor in the world coordinate system at time j
- P_G_Ij represents the position of the inertial sensor at time j in the world coordinate system
- P_G_Ik represents the position of the inertial sensor at time k in the world coordinate system (that is, the actual position of the ground movable platform to be acquired)
- noise_P represents the position noise parameter
- noise_R represents attitude noise parameters.
- the movement speed data obtained in the preceding steps is the movement speed data of at least one wheel of the ground movable platform
- the movement speed data of each wheel can be The processing shown in FIG. 4 can also be used to obtain the average value of the moving speed data of at least one wheel before performing the processing of step S20862 described in FIG. 4, and perform the method shown in FIG. 4 on the average value of the moving speed data. deal with.
- the wheel speed constraint can be further used to process the movement speed data of the movable platform to achieve For the pose recognition of the ground movable platform.
- the method may further include the following steps:
- the second coordinate conversion process may be performed on the movement speed data first to obtain conversion speed data in the world coordinate system, and then the conversion speed data may be processed using the wheel speed constraints to detect the ground Movement information of the movable platform.
- the wheel speed constraint includes a second relationship, and the second relationship is a relationship between the conversion speed data and the actual speed of the ground movable platform.
- the purpose of the aforementioned second coordinate conversion process is to convert the pose data in the wheel speed integral coordinate system W into the pose data in the world coordinate system G.
- the relative posture under the wheel speed integral coordinate system calculated by offline calibration and the inertial coordinate system can be directly used to realize the conversion.
- V_G_Ik R_G_Ik ⁇ R_I_W ⁇ V_Wk+noise_v
- V represents speed
- V_G_Ik represents the speed of the ground movable platform in the world coordinate system at time k (that is: the actual speed of the ground movable platform to be obtained)
- R_G_Ik represents the inertial sensor at time k in the world coordinate system
- R_I_W is used to represent the relative posture between the inertial coordinate system calculated by offline calibration and the wheel speed integral coordinate system
- V_Wk is measured by the aforementioned wheel speed measuring device
- noise_v represents the speed noise parameter.
- R_I_W ⁇ V_Wk is used to characterize the aforementioned second coordinate conversion process.
- the foregoing second relationship is considered by projecting the obtained movement speed data in the wheel speed integral coordinate system to the horizontal direction of the world coordinate system G.
- wheel speed constraints are further added as the iterative solution conditions for nonlinear optimization problems, and the solution is solved, namely The position and posture of the movable platform on the ground can be obtained.
- the embodiment of the present invention uses movement speed data that does not have a vertical component to estimate the pose, and in this process, the pose data is projected to the world coordinate system In the horizontal direction, the error of the uncalibrated inclination angle is extremely small, and it can even be submerged by the noise of the wheel speed itself, which greatly reduces the influence of the uncalibrated inclination angle on the result of pose recognition.
- the uncalibrated inclination angle is A
- the uncalibrated inclination angle is projected to the horizontal direction of the world coordinate system
- the speed and relative position data of the ground movable platform after projection become the original value of cos(A)
- the error ratio is about 1-cos(A).
- the uncalibrated inclination angle in the actual ground mobile platform is generally not large, for example, the uncalibrated inclination angle of the vehicle is generally less than 5°.
- the uncalibrated inclination angle A will bring about sin(A) error in the vertical direction.
- sin(A) is about 0.087, that is It will bring an 8.7% height error to the pose recognition in the vertical direction, and this degree of error will directly lead to the low accuracy of the pose recognition result.
- the overall error ratio between the pose recognition result obtained by the foregoing processing of this solution and the actual pose is less than 0.4%, and this error is only in the horizontal direction, and there is no vertical component. , Thereby reducing as much as possible the influence of components outside the actual constraint direction on the pose recognition result.
- the vehicle has a 0.4% error in the forward direction, which can already be submerged by the noise of the wheel speed itself, which has a very low impact on the actual pose recognition result, and has a relatively small impact on the accuracy of the pose recognition result. small.
- the embodiment of the present invention further considers that in addition to the moving state of the ground movable platform, there may also be a static state.
- the vehicle in addition to the driving state, the vehicle can also be at a standstill.
- a static constraint condition is further provided to restrict the ground movable platform in a static state.
- the stationary constraint is to start when the ground movable platform is at a standstill, and participate in the movement information acquisition process of the ground movable platform.
- the method may also include the following steps: according to the moving speed data, detecting whether the ground movable platform is in a stationary state; The mobile platform is in the stationary state, and the inertial navigation data and the moving speed data are processed by using the stationary constraints of the ground movable platform to detect the movement information of the ground movable platform.
- the movement information of the ground movable platform is detected by using the environmental feature constraints, the inertial feature constraints, the plane feature constraints, and the static constraints to be integrated processing data.
- the method for detecting whether the ground movable platform is in a stationary state is: if the moving speed of the ground movable platform is 0, it is determined that the ground movable platform is in a stationary state.
- Fig. 6 shows a more specific implementation manner. In the method flow shown in Fig. 6, it may further include:
- the so-called static constraint is used to restrict the position and posture of the ground movable platform when it is in a static state.
- the initial pose data and initial velocity data are processed by using the stationary constraint to detect the movement information of the ground movable platform.
- the static constraint includes a third relationship and a fourth relationship; wherein, the third relationship is the relationship between the initial pose data and the actual pose of the ground movable platform, and the fourth relationship is The relationship between the initial speed data and the actual speed of the ground movable platform.
- the essence of the third relationship is that when the ground movable platform is at rest, its position and posture remain unchanged, consistent with the initial pose data and initial velocity data when it is initially at rest, but considering the actual scene
- the third relationship can be specifically expressed as the following two expressions:
- R_G_Ik R_G_I0 ⁇ noise_R0
- P_G_I0 is the initial position of the ground movable platform when it is initially at rest
- R_G_I0 is the initial posture of the ground movable platform when it is initially at rest
- P_G_Ik and R_G_Ik respectively represent the inertial sensor at time k in the world coordinate system
- the position and posture that is, the actual pose data of the ground movable platform to be acquired
- noise_P0 represents the static position noise parameter
- noise_R0 represents the static posture noise parameter.
- the essence of the fourth relationship is that when the ground movable platform is in a stationary state, the moving speed data collected by its wheel speed collecting device is zero. At this time, the static speed noise parameter affects its actual speed. At this time, the fourth relationship can be specifically expressed as the following formula:
- V_G_Ik noise_V0
- V_G_Ik is the speed of the ground movable platform at time k (that is, the actual speed of the ground movable platform to be acquired), and noise_V 0 represents the static speed noise parameter.
- the static constraint can prevent the pose recognition result from drifting slowly due to the interference of other dynamic objects when the movable ground platform is in a static state, and further improve the accuracy of the pose recognition of the movable ground platform in the static state.
- the embodiment of the present invention constructs a nonlinear optimization model for the motion information of the ground movable platform, and directly restricts the speed, position, and attitude of the ground movable platform to obtain the nonlinear optimization model.
- Optimal solution to obtain the movement information of the ground movable platform with higher accuracy which can also significantly improve the reliability of the solution result of the visual inertial navigation method.
- the current frame time can be taken as time k in real time, and the time corresponding to at least one frame before the current frame can be taken as time j, and the foregoing processing can be executed.
- key frames can also be selected based on the aforementioned data, and further, the times corresponding to any two key frames are regarded as time k and time j respectively, and the aforementioned processing is performed.
- At least two key frames may be determined in the at least two frames of image data, and then, according to the at least two The inertial navigation data between key frames is used to obtain the speed data and/or displacement data of the ground movable platform, so that the inertial feature constraints of the ground movable platform are used to process the speed data and/or the displacement data.
- the displacement data to detect the movement information of the ground movable platform.
- determining key frames in at least two frames of images there can be different implementation methods.
- the embodiment of the present invention further provides an embodiment of a device that implements each step and method in the foregoing method embodiment.
- the embodiment of the present invention provides a device for detecting movement information of a ground movable platform. Please refer to FIG. 7.
- the device 700 for detecting movement information of a ground movable platform includes:
- the first acquisition module 71 is configured to acquire at least two frames of image data, the image data including environmental image data of the surrounding environment of the ground movable platform;
- the second acquisition module 72 is used to acquire the inertial navigation data of the ground movable platform
- the detection module 73 is configured to detect the movement information of the ground movable platform according to the image data, the inertial navigation data, and the movement speed data of the ground movable platform;
- the moving speed data is acquired by a wheel speed collecting device mounted on the movable platform on the ground.
- the detection module 73 is specifically used for:
- the movement speed data is processed to detect the movement information of the ground movable platform.
- the plane feature constraint is used to constrain the actual pose of the ground movable platform in the horizontal direction.
- the detection module 73 is specifically used for:
- the plane feature constraint includes a first relationship, and the first relationship is a relationship between the horizontal pose data and the actual pose.
- the detection module 73 is also used for:
- the movement speed data is processed to detect the movement information of the ground movable platform.
- the detection module 73 is specifically used for:
- the wheel speed constraint includes a second relationship, and the second relationship is a relationship between the conversion speed data and the actual speed of the ground movable platform.
- the detection module 73 is also used for:
- the inertial navigation data and the movement speed data are processed by using the stationary constraints of the ground movable platform to detect the movement information of the ground movable platform.
- the detection module 73 is specifically used for:
- the stationary constraint includes a third relationship and a fourth relationship; wherein, the third relationship is the relationship between the initial pose data and the actual pose of the ground movable platform, and the fourth relationship Is the relationship between the initial speed data and the actual speed of the ground movable platform.
- the device 700 for detecting movement information of a ground movable platform may further include:
- the third acquisition module (not shown in Fig. 7) is used to acquire the wheel speed data or the rotation angle data of at least one wheel collected by the wheel speed collection device as the moving speed data.
- the wheel speed collecting device includes at least one of the following: a wheel speed sensor and a wheel encoder.
- the at least one wheel includes: two rear wheels of the ground movable platform.
- the detection module 73 is specifically used for:
- the environmental feature constraints are used to restrict the environment where the ground movable platform is located.
- the detection module 73 is specifically used for:
- the detection module 73 is specifically used for:
- the inertial feature constraint is used to constrain the position and posture of the ground movable platform.
- the detection module 73 is specifically used for:
- the speed data and/or the displacement data are processed by using the inertial feature constraints of the ground movable platform to detect the movement information of the ground movable platform.
- the detection module 73 is specifically used for:
- the speed data and/or the displacement data are processed by using the inertial feature constraints of the ground movable platform to detect the movement information of the ground movable platform.
- the second acquisition module 72 is specifically used for:
- At least one of acceleration and angular velocity data of the ground movable platform collected by an inertial sensor is acquired as the inertial navigation data.
- the ground movable platform includes: a vehicle.
- the device 700 for detecting movement information of a ground movable platform in the embodiment shown in FIG. 7 can be used to implement the technical solutions of the foregoing method embodiments. For its implementation principles and technical effects, you can further refer to the related descriptions in the method embodiments.
- the ground The device 700 for detecting movement information of a movable platform may be a controller or a processor in a movable platform on the ground.
- each module of the ground movable platform motion information detection device 700 shown in FIG. 7 is only a logical function division, and may be fully or partially integrated into a physical entity in actual implementation, or may be physically separated.
- these modules can all be implemented in the form of software called by processing elements; they can also be implemented in the form of hardware; part of the modules can be implemented in the form of software called by the processing elements, and some of the modules can be implemented in the form of hardware.
- the detection module 73 may be a separately established processing element, or it may be integrated in the ground movable platform motion information detection device 700, such as implemented in a certain chip of the terminal. In addition, it may also be stored on the ground in the form of a program.
- a certain processing element of the ground movable platform motion information detection device 700 calls and executes the functions of the above modules.
- the implementation of other modules is similar.
- all or part of these modules can be integrated together or implemented independently.
- the processing element described here may be an integrated circuit with signal processing capability.
- each step of the above method or each of the above modules can be completed by hardware integrated logic circuits in the processor element or instructions in the form of software.
- the above modules may be one or more integrated circuits configured to implement the above methods, such as: one or more application specific integrated circuits (ASIC), or one or more microprocessors (digital singnal processor, DSP), or, one or more field programmable gate arrays (Field Programmable Gate Array, FPGA), etc.
- ASIC application specific integrated circuits
- DSP digital singnal processor
- FPGA Field Programmable Gate Array
- the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processors that can call programs.
- these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC).
- SOC system-on-a-chip
- an embodiment of the present invention provides a device for detecting movement information of a ground movable platform. Please refer to FIG. 8.
- the device 700 for detecting movement information of a ground movable platform includes:
- the computer program is stored in the memory 710 and is configured to be executed by the processor 720 to implement the method described in the foregoing embodiment.
- the number of processors 720 in the device 700 for detecting movement information of a ground movable platform may be one or more, and the processors 720 may also be referred to as a processing unit, which may implement certain control functions.
- the processor 720 may be a general-purpose processor or a special-purpose processor.
- the processor 720 may also store instructions, and the instructions may be executed by the processor 720, so that the device 700 for detecting movement information of a ground movable platform executes the method described in the foregoing method embodiment. method.
- the device 700 for detecting movement information of a ground movable platform may include a circuit, which may implement the sending or receiving or communication functions in the foregoing method embodiments.
- the number of memories 710 in the device 700 for detecting movement information of a ground movable platform may be one or more, and instructions or intermediate data are stored in the memory 710, and the instructions may be executed on the processor 720 , So that the device 700 for detecting movement information of a ground movable platform executes the method described in the foregoing method embodiment.
- other related data may also be stored in the memory 710.
- instructions and/or data may also be stored in the processor 720.
- the processor 720 and the memory 710 may be provided separately or integrated together.
- a transceiver 730 is also provided in the device 700 for detecting movement information of a ground movable platform.
- the transceiver 730 may be called a transceiver unit, a transceiver, a transceiver circuit, or a transceiver, etc. , Used for data transmission or communication with test equipment or other terminal equipment, and will not be repeated here.
- the memory 710, the processor 720, and the transceiver 730 are connected and communicate via a bus.
- the transceiver 730 can release the tested package to each test terminal, and the transceiver 730 can also be used to receive each test. Test run data fed back by the terminal.
- the processor 720 is configured to complete corresponding determination or control operations, and optionally, may also store corresponding instructions in the memory 710. For the specific processing manner of each component, reference may be made to the related description of the foregoing embodiment.
- the ground movable platform motion information detection device 700 shown in FIG. 7 or FIG. 8 may be a separate device, for example, may be one of the auxiliary driving devices (or automatic driving devices) of the vehicle. Separately installable processor or processing device; or, the ground movable platform motion information detection device 700 may be integrated with other devices in the ground movable platform, for example, may be integrated in a supercomputer platform or a storage system.
- the device 700 for detecting movement information of a ground movable platform may only include a processor, so as to implement the violation of the law as described in any implementation manner of the first embodiment, or, In addition, it can further include devices such as vision sensors.
- the device 700 for detecting movement information of the ground movable platform generally can directly call the data collected by the wheel speed collection device, inertial sensor, camera, etc. carried on the ground movable platform.
- the ground movable platform motion information detection device 700 can communicate with other data collection equipment carried on the ground movable platform through a bus or wireless communication, so as to obtain various data collected by these data collection equipment .
- an embodiment of the present invention provides a readable storage medium having a computer program stored thereon, and the computer program is executed by a processor to implement the method as described in the first embodiment.
- an embodiment of the present invention provides a system for detecting movement information of a ground movable platform. Please refer to FIG. 9.
- the system for detecting movement information of a ground movable platform 900 includes:
- the image acquisition device 910 is configured to acquire at least two frames of the image data
- An inertial sensor 920 configured to acquire the inertial navigation data of the ground movable platform
- Wheel speed collecting device 930 used to obtain the moving speed data of the ground movable platform
- the device 700 for detecting movement information of a ground movable platform is used for:
- the moving speed data is acquired by the wheel speed collecting device 930 mounted on the movable platform on the ground.
- the device 700 for detecting movement information of a ground movable platform is specifically used for:
- the movement speed data is processed to detect the movement information of the ground movable platform.
- the plane feature constraint is used to constrain the actual pose of the ground movable platform in the horizontal direction.
- the device 700 for detecting movement information of a ground movable platform is specifically used for:
- the plane feature constraint includes a first relationship, and the first relationship is a relationship between the horizontal pose data and the actual pose.
- the device 700 for detecting movement information of a ground movable platform is further used for:
- the movement speed data is processed to detect the movement information of the ground movable platform.
- the device 700 for detecting movement information of a ground movable platform is specifically used for:
- the wheel speed constraint includes a second relationship, and the second relationship is a relationship between the conversion speed data and the actual speed of the ground movable platform.
- the device 700 for detecting movement information of a ground movable platform is further used for:
- the inertial navigation data and the movement speed data are processed by using the stationary constraints of the ground movable platform to detect the movement information of the ground movable platform.
- the device 700 for detecting movement information of a ground movable platform is specifically used for:
- the stationary constraint includes a third relationship and a fourth relationship; wherein, the third relationship is the relationship between the initial pose data and the actual pose of the ground movable platform, and the fourth relationship Is the relationship between the initial speed data and the actual speed of the ground movable platform.
- the device 700 for detecting movement information of a ground movable platform is further used for:
- the wheel speed data or the turning angle data of at least one wheel collected by the wheel speed collecting device 930 is acquired as the moving speed data.
- the wheel speed collecting device 930 includes at least one of the following: a wheel speed sensor and a wheel encoder.
- the at least one wheel includes: two rear wheels of the ground movable platform.
- the device 700 for detecting movement information of a ground movable platform is specifically used for:
- the environmental feature constraints are used to restrict the environment where the ground movable platform is located.
- the device 700 for detecting movement information of a ground movable platform is specifically used for:
- the device 700 for detecting movement information of a ground movable platform is specifically used for:
- the inertial feature constraint is used to constrain the position and posture of the ground movable platform.
- the device 700 for detecting movement information of a ground movable platform is specifically used for:
- the speed data and/or the displacement data are processed by using the inertial feature constraints of the ground movable platform to detect the movement information of the ground movable platform.
- the device 700 for detecting movement information of a ground movable platform is specifically used for:
- the speed data and/or the displacement data are processed by using the inertial feature constraints of the ground movable platform to detect the movement information of the ground movable platform.
- the device 700 for detecting movement information of a ground movable platform is specifically used for:
- At least one of the acceleration and angular velocity data of the ground movable platform collected by the inertial sensor 920 is acquired as the inertial navigation data.
- the ground movable platform includes a vehicle.
- the ground movable platform motion information detection device 700 may interact and transmit data through wired or wireless communication.
- the image acquisition device 910 may interact and transmit data through wired or wireless communication.
- the inertial sensor 920 may interact and transmit data through wired or wireless communication.
- the ground movable platform motion information detection system 900 may be specifically an ADAS system carried in a vehicle.
- an embodiment of the present invention provides a ground movable platform, please refer to FIG. 10, the ground movable platform includes:
- the power system 1020 is installed on the fuselage 1010 to provide operating power
- the device 700 for detecting movement information of a ground movable platform is configured to execute the method described in any implementation manner of the embodiment;
- the image acquisition device 910 is configured to acquire at least two frames of the image data
- An inertial sensor 920 configured to acquire the inertial navigation data of the ground movable platform
- the wheel speed collecting device 930 is used to obtain the moving speed data of the ground movable platform.
- the embodiment of the present invention is not limited to the specific structure and composition of the fuselage 1010 and the power system 1020 in the ground movable platform, and will not be repeated here.
- the ground movable platform involved in the embodiment of the present invention may include, but is not limited to, a vehicle.
- other ground machinery devices may also be included, where the ground machinery devices may include, but are not limited to: at least one of ground machinery toys and ground robots.
- the ground machine toy may be a toy car, and for example, the ground robot may be a ground sweeping robot, a ground dispatching robot, and so on.
- a person of ordinary skill in the art can understand that all or part of the steps in the above method embodiments can be implemented by a program instructing relevant hardware.
- the foregoing program can be stored in a computer readable storage medium. When the program is executed, it is executed. Including the steps of the foregoing method embodiment; and the foregoing storage medium includes: ROM, RAM, magnetic disk, or optical disk and other media that can store program codes.
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Aviation & Aerospace Engineering (AREA)
- Automation & Control Theory (AREA)
- Navigation (AREA)
Abstract
本发明提供一种地面可移动平台与其运动信息检测方法、系统,该方法包括:获取至少两帧图像数据,所述图像数据包括所述地面可移动平台周围环境的环境图像数据,然后,获取所述地面可移动平台的惯性导航数据,从而,根据所述图像数据、所述惯性导航数据与所述地面可移动平台的移动速度数据,检测所述地面可移动平台的运动信息;其中,所述移动速度数据通过搭载于所述地面可移动平台的轮速采集装置获取。本方面实施例所提供的方法能够提高位姿识别结果的精确度与可靠性。
Description
本申请涉及智能驾驶技术领域,尤其涉及一种地面可移动平台与其运动信息检测方法、系统。
车辆自身的位置、姿态识别,是自动驾驶的关键技术之一,其目的是利用车上搭载的传感器,识别出车辆自身的位置以及姿态。而车辆自身位置、姿态识别结果的精度和可靠性,直接决定了后续传感器数据处理、车辆导航和控制的精度和可靠性。如何兼顾高精度与高可靠性的位姿识别,就成为本领域重点关注的技术问题。
目前,基于视觉同步定位与地图构建(SLAM,Simultaneous Localization And Mapping)、视觉惯性导航系统(VINS,Visual Inertial Navigtion System)及相关数据融合的方法作为一种常用的位姿识别方法,假设了车辆的速度方向与惯性传感器(Inertial measurement unit,IMU)所指车头方向平行,并在此基础上,检测和跟踪图像中的图像特征点来恢复三维运动信息,从而,得到车辆的位姿识别结果。
但是,在实际的车辆行驶过程中,车辆中的惯性传感器会由于姿态漂移以及视觉上的环境误差,导致位姿识别结果可能存在实际约束方向外的分量,使得位姿识别结果存在识别精度与可靠性较差的问题。
发明内容
本发明实施例提供一种地面可移动平台与其运动信息检测方法、系统,用以提高位姿识别结果的精确度与可靠性。
第一方面,本发明实施例提供一种地面可移动平台运动信息检测方法,包括:
获取至少两帧图像数据,所述图像数据包括所述地面可移动平台周围环境的环境图像数据;
获取所述地面可移动平台的惯性导航数据;
根据所述图像数据、所述惯性导航数据与所述地面可移动平台的移动速度数据,检测所述地面可移动平台的运动信息;
其中,所述移动速度数据通过搭载于所述地面可移动平台的轮速采集装置获取。
第二方面,本发明实施例提供一种地面可移动平台运动信息检测装置,包括:
存储器;
处理器;以及
计算机程序;
其中,所述计算机程序存储在所述存储器中,并被配置为由所述处理器执行以实现如第一方面所述的方法。
第三方面,本发明实施例提供一种计算机可读存储介质,其上存储有计算机程序,
所述计算机程序被处理器执行以实现如第一方面所述的方法。
第四方面,本发明实施例提供一种地面可移动平台运动信息检测系统,包括:
图像采集装置,用于获取至少两帧所述图像数据;
惯性传感器,用于获取所述地面可移动平台的所述惯性导航数据;
轮速采集装置,用于获取所述地面可移动平台的移动速度数据;
地面可移动平台运动信息检测装置,用于:
获取所述图像采集装置获取到的至少两帧所述图像数据,所述图像数据包括所述地面可移动平台周围环境的环境图像数据;
获取所述惯性传感器获取到的所述惯性导航数据;
根据所述图像数据、所述惯性导航数据与所述地面可移动平台的移动速度数据,检测所述地面可移动平台的运动信息;
其中,所述移动速度数据通过搭载于所述地面可移动平台的所述轮速采集装置获取。
第五方面,本发明实施例提供一种地面可移动平台,包括:
机身;
动力系统,安装于所述机身,用于提供运行动力;
地面可移动平台运动信息检测装置,用于执行如第一方面所述的方法;
图像采集装置,用于获取至少两帧所述图像数据;
惯性传感器,用于获取所述地面可移动平台的所述惯性导航数据;
轮速采集装置,用于获取所述地面可移动平台的移动速度数据。
本发明实施例提供的地面可移动平台与其运动信息检测方法、系统,通过图像数据、所述惯性导航数据与移动速度数据,来检测地面可移动平台的运动功能信息,由于移动速度数据是约束在路面上的水平方向数据,不存在垂直分量,因此,本方案通过移动速度数据来对地面可移动平台的运动信息进行水平方向上的限制,避免由于IMU的姿态漂移导致运动信息出现实际约束方向外的分量的情况以及该情况对于位姿识别结果的影响,这有利于提高识别精度与识别结果的可靠性。
图1为本发明实施例所涉及到的未标定倾角示意图;
图2为本发明实施例所提供的一种地面可移动平台运动信息检测方法的流程示意图;
图3为本发明实施例所提供的另一种地面可移动平台运动信息检测方法的流程示意图;
图4为本发明实施例所提供的另一种地面可移动平台运动信息检测方法的流程示意图;
图5为本发明实施例所提供的另一种地面可移动平台运动信息检测方法的流程示意图;
图6为本发明实施例所提供的另一种地面可移动平台运动信息检测方法的流程示意图;
图7为本发明实施例所提供的一种地面可移动平台运动信息检测装置的功能方块图;
图8为本发明实施例所提供的一种地面可移动平台运动信息检测装置的实体结构示意图;
图9为本发明实施例所提供的一种地面可移动平台运动信息检测系统的 架构示意图;
图10为本发明实施例所提供的一种地面可移动平台的架构示意图。
这里将详细地对示例性实施例进行说明,其示例表示在附图中。下面的描述涉及附图时,除非另有表示,不同附图中的相同数字表示相同或相似的要素。以下示例性实施例中所描述的实施方式并不代表与本公开相一致的所有实施方式。相反,它们仅是与如所附权利要求书中所详述的、本公开的一些方面相一致的装置和方法的例子。
本发明具体的应用场景为针对地面可移动平台的位置、姿态识别场景。更进一步的,可以为针对地面可移动平台的自动驾驶或自动控制系统。
其中,本发明实施例所涉及到的地面可移动平台,可以包括但不限于:车辆。除此之外,还可以包括其他地面机器装置,其中,地面机器装置可以包括但不限于:地面机器玩具与地面机器人中的至少一种。例如,地面机器玩具可以为智能玩具汽车,又例如,地面机器人可以为地面扫地机器人、地面调度机器人等。
为了便于说明,本发明实施例以车辆为例进行具体说明。
而如本发明前述背景技术所述,现有的基于SLAM、VINS与相关数据融合实现位姿识别的方法是以假设的车辆的速度方向与IMU所指的车头方向平行为基础实现的。然而,在车辆的实际行驶过程中,车辆的速度方向(车轮坐标系)与车头方向(惯性坐标系)存在垂直方向上的误差,也就是未标定倾角误差。
以车辆为例,车辆中设置有悬挂系统,例如,车轮与车体之间并不是刚性连接的。特别地,当车上的重心前后位置不同,会导致车体与路面之间的倾角存在差异。图1示出了车辆的未标定倾角误差的示意图。如图1所示,当车辆的重心靠后时,车辆的车头会向上;当车辆的重心靠前时,车辆的车头会向下。同时,在行驶过程中,车体与路面之间的倾角也会发生振荡。这两部分综合,会导致惯性传感器所指示的惯性坐标系与车轮坐标系之间存在未标定倾角误差,从而,位于惯性坐标系下的车头方向(图1中实线箭头方向)与位于车轮坐标系下的速度方向(图1中虚线箭头方向)就存在未标定 倾角误差。
而基于如图1所示的未标定倾角的误差,使得前述假设与实际识别情况存在不符,从而位姿识别结果存在实际约束方向外的分量,进而导致了位姿识别结果存在识别精度与可靠性较差的问题。
本发明提供的技术方案,旨在解决现有技术的如上技术问题。
下面以具体地实施例对本发明的技术方案以及本申请的技术方案如何解决上述技术问题进行详细说明。下面这几个具体的实施例可以相互结合,对于相同或相似的概念或过程可能在某些实施例中不再赘述。下面将结合附图,对本发明的实施例进行描述。
实施例一
本发明实施例提供了一种地面可移动平台运动信息检测方法。请参考图2,该方法包括如下步骤:
S202,获取至少两帧图像数据,所述图像数据包括所述地面可移动平台周围环境的环境图像数据。
其中,惯性导航数据可以通过地面可移动平台上设置的图像采集装置获取得到。图像采集装置可以包括但不限于:摄像头,其中,摄像头可以包括但不限于:彩色摄像头、灰度摄像头等。以及,本发明实施例对于图像采集装置的具体参数无特别限定。例如,利用摄像头采集图像数据时,摄像头的数目可以为一个或多个,可以为标定好的一个或一对双目摄像头。
S204,获取所述地面可移动平台的惯性导航数据。
其中,惯性导航数据可以通过地面可移动平台上设置的惯性传感器(Inertial Measurement Unit,IMU)测量得到。其数据可以具体包括但不限于:加速度和角速度中的至少一种。
如此,该步骤在实现时,仅需要获取惯性传感器采集得到的所述地面可移动平台的加速度与角速度数据中的至少一种,以作为所述惯性导航数据即可。
S206,通过搭载于所述地面可移动平台的轮速采集装置,获取所述地面可移动平台的移动速度数据。
此处,为了便于说明,示出了获取轮速采集装置采集到的移动速度数据的步骤。
本发明实施例所涉及到的轮速采集装置可以包括但不限于如下至少一种:轮速传感器、轮子编码器。需要理解的是,本发明实施例所涉及的轮速采集装置不仅限于轮式驱动地面可移动平台的移动速度测量,也包括其他方式驱动地面可移动品台的移动速度测量,例如履带式机器人等。
基于此,本发明实施例所涉及到的移动速度数据可以包括但不限于:所述地面可移动平台的至少一个轮子的轮速数据或转角数据。可知,在执行S208所述步骤之前,本发明实施例还可以包括如下步骤:获取所述轮速采集装置采集到的至少一个轮子的轮速数据或转角数据,以作为所述移动速度数据。
若地面可移动平台为四轮车辆,在一种实现方式中,前述至少一个轮子可以包括但不限于:所述地面可移动平台的两个后轮。也就是,可以获取地面可移动平台的两个后轮(即左后轮和右后轮)的轮速数据,以作为所述移动速度数据。
但本发明实施例对于地面可移动平台的至少一个轮子是哪个轮子无特别限定。除前述方式以外,在其他的实现方式中,也可以获取地面可移动平台的四个轮子的轮速数据或转角数据,以作为所述移动速度数据;或者,也可以获取地面可移动平台的任意一个前轮的转角数据作为所述移动速度数据。在实际实现场景中,可以根据实际需要对轮子进行预设,不再赘述。
S208,根据所述图像数据、所述惯性导航数据与所述地面可移动平台的移动速度数据,检测所述地面可移动平台的运动信息。
基于前述采集到的数据,检测所述地面可移动平台在世界坐标系下的位置数据、姿态数据与速度数据。
在执行该步骤时,除考虑所述移动速度数据之外,还可以进一步考虑所述图像数据与所述惯性导航数据,来实现对所述地面可移动平台的运动信息的检测。
为了便于说明,以下,分别针对前述3种数据的处理方式进行说明。可知,以下说明是在前述发明构思的前提下实现的,即便是针对某一种数据的处理来实现运动信息的检测,也不意味着不考虑其他数据。例如,后续提及“根据所述移动速度数据,检测所述地面可移动平台的运动信息”是指针对移动速度数据进行处理以检测运动信息的处理方式,并不意味着单独依靠该数据实现对所述地面可移动平台的运动信息的检测。
具体而言,S208在实现时可以参考图3所示的流程。如图3所示的实现方案中,是根据所述图像数据、所述惯性导航数据与所述移动速度数据,检测所述地面可移动平台的运动信息的一种可能的设计。如图3所示,S208包括如下步骤:
S2082,利用所述地面可移动平台的环境特征约束,处理所述图像数据,以检测所述地面可移动平台的运动信息。
其中,所述环境特征约束用于约束所述地面可移动平台所在环境。具体的,不同时刻的所述图像数据之间的对应关系,即在相近时刻内的图像数据中的环境具有重合的部分,因此,环境特征约束可实现对地面可移动平台的环境情况的约束。具体而言,其可通过至少两帧图像数据之间的特征点来实现环境特征约束,其中,特征点为所述地面可移动平台所在环境的特征点。
S2084,利用所述地面可移动平台的惯性特征约束,处理所述惯性导航数据,以检测所述地面可移动平台的运动信息。
其中,所述惯性特征约束用于约束所述地面可移动平台的位置和姿态。惯性传感器采集到的惯性导航数据,可以为地面可移动平台的姿态提供惯性特征约束。
S2086,利用所述地面可移动平台的平面特征约束,处理所述移动速度数据,以检测所述地面可移动平台的运动信息。
其中,所述平面特征约束用于约束所述地面可移动平台在水平方向上的实际位姿。
如前所述,地面可移动平台的移动速度数据是约束在路面上的水平方向数据,不存在垂直分量,因此,本发明实施例通过移动速度数据来对地面可移动平台的运动信息进行水平方向上的限制,避免由于IMU的姿态漂移导致运动信息出现垂直方向上分量(也即,实际约束方向外的分量)的情况以及该情况对于位姿识别结果的影响,从而,达到提高识别精度与识别结果的可靠性的效果。
S2088,得到所述地面可移动平台的运动信息。
前述环境特征约束、惯性特征约束与平面特征约束可以作为非线性优化问题的迭代求解条件,如此,对非线性优化问题进行求解,即可得到地面可移动平台的位置、姿态等运动信息。
以下,对图3所示的各约束条件可以通过如图4所示的方式实现:
首先,针对S2082中利用环境特征约束处理图像数据的步骤,如图4所示,可以通过如下方式来实现:
S20822,对所述至少两帧图像数据进行特征点匹配,得到环境特征点。
具体而言,就是提取各帧图像数据的特征点,然后,对各帧图像的特征点进行匹配,若在至少两帧图像中能够匹配到相同的特征点,这些特征点可用于表征地面可移动平台所在环境的环境特征,此时,将匹配到的特征点作为环境特征点。
S20824,利用所述环境特征约束处理所述环境特征点,以检测所述地面可移动平台的运动信息。
具体而言,可利用前述得到的环境特征点,来检测地面可移动平台的运动信息。
其次,针对S2084中利用惯性特征约束处理惯性导航数据的步骤,如图4所示,可以通过如下方式来实现:
S20842,根据所述惯性导航数据,获取所述地面可移动平台的速度数据和/或位移数据。
具体而言,对所述惯性导航数据进行处理,以得到所述地面可移动平台的速度数据和/或位移数据。其中,处理方式可以包括但不限于:积分算法。
S20844,利用所述地面可移动平台的惯性特征约束,处理所述速度数据和/或所述位移数据,以检测所述地面可移动平台的运动信息。
该步骤中,可利用前述得到的速度数据和/或位移数据,来检测所述地面可移动平台的运动信息。
再者,针对S2086中利用平面特征约束处理移动速度数据的步骤,可通过如图4所示的方式获取得到。如图4所述,S2086可以通过如下方式实现:
S20862,根据所述移动速度数据,获取所述地面可移动平台的位姿数据。
具体而言,对所述移动速度数据进行处理,以得到所述地面可移动平台的位姿数据。其中,对移动速度进行处理时,可以针对任意两个时刻之间的移动速度数据进行处理,以根据这两个时刻之间的移动速度情况,得到地面可移动平台的位姿数据。后续,为便于说明,假设地面可移动平台的位姿数据是通过对第一时刻(后续简称为k时刻)与第二时刻(后续简称为j时刻) 之间的移动速度数据进行处理得到的。而经过该步骤处理之后,得到的地面可移动平台的位姿数据可以为:P_Wk_Wj和R_Wk_Wj;其中,P表示位置,R表示姿态,W表示轮速积分坐标系,而P_Wk_Wj表示地面可移动平台在k时刻与j时刻之间在轮速积分坐标系上的相对位置,R_Wk_Wj表示地面可移动平台在k时刻与j时刻之间在轮速积分坐标系上的相对姿态。
S20864,对所述位姿数据进行第一坐标转换处理,得到惯性坐标系下的惯性位姿数据。
如前所述,S20862所得到的位姿数据为轮速积分坐标系下的相对位姿,而在具体对地面可移动平台的位姿进行识别时,需要获取到的是某一时刻(本发明实施例假设为k时刻)惯性传感器在世界坐标系下的位置、姿态与速度数据。
由此,需要对前述步骤获取到的地面可移动平台的位姿数据P_Wk_Wj和R_Wk_Wj进行坐标转换,使其由轮速积分坐标系下的相对位姿转换为惯性坐标系下的相对位姿。假设惯性坐标系由I进行表示,则该步骤在实现时,需要获取的数据为:R_Ik_Ij和P_Ik_Ij。其中,R_Ik_Ij表示地面可移动平台在k时刻与j时刻之间在惯性坐标系上的相对姿态,P_Ik_Ij表示地面可移动平台在k时刻与j时刻之间在惯性坐标系上的相对位置。
该坐标转换步骤可以按照如下式所示的方式实现:
R_Ik_Ij=R_I_W×R_Wk_Wj×R_W_I
P_Ik_Ij=P_I_W+R_I_W×P_Wk_Wj–R_Ik_Ij×P_I_W
其中,R_I_W用于表示离线标定计算的惯性坐标系与轮速积分坐标系之间的相对姿态,而R_W_I是R_I_W的逆矩阵;P_I_W用于表示离线标定计算的惯性坐标系与轮速积分坐标系之间的相对位置,而P_W_I是P_I_W的逆矩阵。
S20866,获取所述惯性位姿数据的水平分量,得到所述惯性坐标系下的水平位姿数据。
考虑到世界坐标系的约束具备六自由度,因此,在具体获取平面特征约束时,仅需要考虑这些数据在水平方向上的分量,以获取平面特征约束即可。该步骤在具体实现时,可以通过将前述惯性位姿数据投影到水平方向上来实现。
针对前述获取到的相对位置数据,取其水平分量即可;而针对前述获取到的相对姿态数据,则只去其z轴分量(水平分量)即可。为了便于说明,将该步骤得到的惯性坐标系下的水平位姿数据记录为:R_Ik_Ij_H和P_Ik_Ij_H,其中,H表示水平分量,R_Ik_Ij_H为R_Ik_Ij的水平分量,P_Ik_Ij_H为P_Ik_Ij的水平分量。
S20868,利用所述平面特征约束处理所述水平位姿数据,以检测所述地面可移动平台的运动信息。
其中,所述平面特征约束包含第一关系,而所述第一关系为所述水平位姿数据与所述实际位姿之间的关系。在世界坐标系G下,所述第一关系可以表征为下式:
R_G_Ik×P_Ik_Ij_H=P_G_Ij-P_G_Ik+noise_P
R_Ik_Ij_H=R_G_Ik^(-1)×R_G_Ij×noise_R
其中,G表示世界坐标系,R_G_Ik表示k时刻惯性传感器在世界坐标系下的姿态(也即:待获取的地面可移动平台的实际姿态),R_G_Ij表示j时刻惯性传感器在世界坐标系下的姿态,P_G_Ij表示j时刻惯性传感器在世界坐标系下的位置,P_G_Ik表示k时刻惯性传感器在世界坐标系下的位置(也即:待获取的地面可移动平台的实际位置),noise_P表示位置噪声参数,noise_R表示姿态噪声参数。
此外,还需要说明的是,由于前述步骤中获取到的移动速度数据为地面可移动平台的至少一个轮子的移动速度数据,因此,在执行S20862步骤时,可以对每个轮子的移动速度数据进行如图4所示的处理,亦可以在执行图4所述的S20862步骤的处理之前,获取至少一个轮子的移动速度数据的平均值,并对移动速度数据的平均值执行如图4所示的处理。
在另一种设计中,在执行如图2所示的S208步骤,或图3、图4所述的S2088步骤之前,还可以进一步利用轮速约束来处理可移动平台的移动速度数据,来实现针对地面可移动平台的位姿识别。
此时,可以参考图5,该方法还可以包括如下步骤:
S2085,利用所述地面可移动平台的轮速约束,处理所述移动速度数据,以检测所述地面可移动平台的运动信息。
执行该步骤时,可以首先对所述移动速度数据进行第二坐标转换处理, 得到世界坐标系下的转换速度数据,然后,利用所述轮速约束处理所述转换速度数据,以检测所述地面可移动平台的运动信息。其中,所述轮速约束包括第二关系,所述第二关系为所述转换速度数据与所述地面可移动平台的实际速度之间的关系。
其中,前述第二坐标转换处理的目的在于,将轮速积分坐标系W下的位姿数据转换为世界坐标系G下的位姿数据。此时,可以直接利用离线标定计算的轮速积分坐标系与惯性坐标系之下的相对姿态来实现该转换。
此时,第二关系可以具体表示为下式:
V_G_Ik=R_G_Ik×R_I_W×V_Wk+noise_v
其中,V表示速度,V_G_Ik表示地面可移动平台在k时刻的世界坐标系下的速度(也即:待获取的地面可移动平台的实际速度),而R_G_Ik表示k时刻惯性传感器在世界坐标系下的姿态(也即:待获取的地面可移动平台的实际姿态),R_I_W用于表示离线标定计算的惯性坐标系与轮速积分坐标系之间的相对姿态,V_Wk为前述轮速测量装置测量得到的k时刻的移动速度数据,noise_v表示速度噪声参数。此外,R_I_W×V_Wk用于表征前述第二坐标转换处理的过程。
前述第二关系是将前述获取到的轮速积分坐标系下的移动速度数据投影到世界坐标系G的水平方向进行考虑的。
基于如图5所示的设计,是在利用前述环境特征约束、惯性特征约束与平面特征约束的基础上,进一步增加轮速约束条件,作为非线性优化问题的迭代求解条件,并进行求解,即可得到地面可移动平台的位置、姿态等运动信息。
通过前述图2-图5所示的所设计,本发明实施例利用不具备垂直方向分量的移动速度数据来进行位姿估计,并且,在此过程中,将位姿数据投影到世界坐标系的水平方向上,那么,未标定倾角的误差极小,甚至可以被轮速本身的噪声淹没,极大地降低了未标定倾角对位姿识别结果的影响。
具体而言,假设未标定倾角的大小为A,那么,将未标定倾角投影到世界坐标系的水平方向,那么,投影后地面可移动平台的速度和相对位置数据成为原来数值的cos(A),带来的误差比例约为1-cos(A)。
而在实际的地面可移动平台中的未标定倾角一般不大,例如,车辆的未 标定倾角一般小于5°。
若如现有技术的处理方式直接进行位姿识别,则未标定倾角A将为垂直方向上带来sin(A)的误差,在A=5°时sin(A)约为0.087,也就是,将给垂直方向上的位姿识别带来8.7%的高度误差,这种程度的误差将直接导致位姿识别结果的低准确率。
相比之下,通过本方案的前述处理而得到的位姿识别结果,与实际位姿之间的综合误差比例小于0.4%,并且这一误差只在水平方向上,不存在垂直方向上的分量,从而尽可能地降低了实际约束方向外的分量对位姿识别结果的影响。而在实际的位姿识别中,车辆前进方向上具备0.4%的误差,已经可以被轮速本身的噪声淹没,对实际位姿识别结果的影响极低,对位姿识别结果的准确率影响较小。
除前述设计之外,本发明实施例还进一步考虑到所述地面可移动平台除运动这一种状态外,还可能存在一种静止状态。例如,车辆除行驶状态外,还可以处于静止状态。
本发明实施例中,为了更精确地识别地面可移动平台的位姿,还进一步提供了一种静止约束条件,以对处于静止状态的地面可移动平台进行约束。其中,静止约束是在可以地面可移动平台处于静止状态下启动,并参与地面可移动平台的运动信息获取过程。
此时,该方法除包含前述各实现方式所述的步骤之外,该方法还可以包括如下步骤:根据所述移动速度数据,检测所述地面可移动平台是否处于静止状态;若所述地面可移动平台处于所述静止状态,利用所述地面可移动平台的静止约束处理所述惯性导航数据与所述移动速度数据,以检测所述地面可移动平台的运动信息。在这种设计方案中,地面可移动平台的运动信息是利用所述环境特征约束、所述惯性特征约束、所述平面特征约束、所述静止约束综合处理数据检测得到的。
其中,检测地面可移动平台是否处于静止状态的方法为:若地面可移动平台的移动速度为0,则确定地面可移动平台处于静止状态。
图6示出了一种更具体的实现方式,在如图6所示的方法流程中,还可以包括:
S2087,当所述移动速度数据为零时,利用所述地面可移动平台的静止约 束所述惯性导航数据与所述移动速度数据,以检测所述地面可移动平台的运动信息。
可知,若移动速度数据不为零,则可以按照如图4或图5所示流程,利用所述环境特征约束、所述惯性特征约束、所述平面特征约束(还可以包括轮速约束)进行优化求解,来获取到所述地面可移动平台的运动信息。不再赘述。
具体而言,所谓静止约束,用于对地面可移动平台处于静止状态时的位置和姿态保持不变进行约束。
具体而言,可以通过如下方式获取得到:
根据所述惯性导航数据与所述移动速度数据,获取所述地面可移动平台在初始处于静止状态时的初始位姿数据与初始速度数据;
利用所述静止约束处理所述初始位姿数据与初始速度数据,以检测所述地面可移动平台的运动信息。
其中,静止约束包括第三关系与第四关系;其中,所述第三关系为所述初始位姿数据与所述地面可移动平台的实际位姿之间的关系,所述第四关系为所述初始速度数据与所述地面可移动平台的实际速度之间的关系。
一方面,第三关系的实质在于:地面可移动平台处于静止状态时,其位置和姿态保持不变,与初始处于静止状态时的初始位姿数据和初始速度数据保持一致,但考虑到实际场景中的噪声影响,此时,第三关系可以具体表示为如下两个表达式:
P_G_Ik=P_G_I0+noise_P0
R_G_Ik=R_G_I0×noise_R0
其中,P_G_I0为地面可移动平台在初始处于静止状态时的初始位置,R_G_I0为地面可移动平台在初始处于静止状态时的初始姿态,而P_G_Ik和R_G_Ik则分别表示k时刻惯性传感器在世界坐标系下的位置和姿态(也即:待获取的地面可移动平台的实际位姿数据),noise_P0表示静止位置噪声参数,noise_R0表示静止姿态噪声参数。
另一方面,第四关系的实质在于:地面可移动平台处于静止状态时,其轮速采集装置采集到的移动速度数据为零,此时,对其实际速度产生影响的为静止速度噪声参数,此时,第四关系可以具体表示为下式:
V_G_Ik=noise_V0
其中,V_G_Ik为地面可移动平台在k时刻时的速度(也即:待获取的地面可移动平台的实际速度),noise_V 0表示静止速度噪声参数。
通过静止约束,可以防止地面可移动平台在处于静止状态时由于其他动态物体的干扰而导致位姿识别结果出现缓慢漂移的现象,进一步提高了静止状态下的地面可移动平台的位姿识别准确率。
通过前述方案,本发明实施例为地面可移动平台的运动信息构建了非线性优化模型,并对地面可移动平台的速度、位置与姿态分别进行直接的条件约束,来获取该非线性优化模型的最优解,以得到精确度较高的地面可移动平台的运动信息,这也可以显著地提高视觉惯性导航方法求解结果的可靠性。
此外,还需要说明的是,本发明实施例所提供的技术方案,对于前述k时刻与j时刻的选取无特别限定。
在一种实际实现本方案过程中,可以实时的将当前帧时刻作为k时刻,将当前帧之前的至少一帧对应的时刻作为j时刻,执行前述处理。
在一种实际实现本方案过程中,还可以基于前述得到的数据,进行关键帧的选取,进而,将任意两个关键帧对应的时刻分别作为k时刻和j时刻,执行前述处理。
以图3所述流程为例,在执行S2084与S2086之前,还需要在所述至少两帧图像数据中确定至少两个关键帧,然后,分别利用环境特征约束、惯性特征约束与水平特征约束来处理所述至少两个关键帧之间的数据,以检测所述地面可移动平台的运动信息。
以图3所述流程中利用惯性特征约束对惯性导航数据进行处理的步骤为例,此时,可以在所述至少两帧图像数据中确定至少两个关键帧,然后,根据所述至少两个关键帧之间的所述惯性导航数据,获取所述地面可移动平台的速度数据和/或位移数据,从而,利用所述地面可移动平台的惯性特征约束,处理所述速度数据和/或所述位移数据,以检测所述地面可移动平台的运动信息。
而在至少两帧图像中确定关键帧,则可以有不同的实现方式,实际实现时,可以根据需要自定义设计取关键帧的规则。例如,可以设置取关键帧的周期,在每隔5帧取一帧为关键帧;又例如,可以在连续的N(N为大于1 的整数)帧中随机选择一帧作为关键帧;又例如,可以通过前述针对至少两帧的图像特征点的匹配结果,将环境特征点匹配程度较高的至少两帧来作为关键帧,不再赘述。
可以理解的是,上述实施例中的部分或全部步骤或操作仅是示例,本申请实施例还可以执行其它操作或者各种操作的变形。此外,各个步骤可以按照上述实施例呈现的不同的顺序来执行,并且有可能并非要执行上述实施例中的全部操作。
实施例二
基于上述实施例一所提供的地面可移动平台运动信息检测方法,本发明实施例进一步给出实现上述方法实施例中各步骤及方法的装置实施例。
本发明实施例提供了一种地面可移动平台运动信息检测装置,请参考图7,该地面可移动平台运动信息检测装置700,包括:
第一获取模块71,用于获取至少两帧图像数据,所述图像数据包括所述地面可移动平台周围环境的环境图像数据;
第二位获取模块72,用于获取所述地面可移动平台的惯性导航数据;
检测模块73,用于根据所述图像数据、所述惯性导航数据与所述地面可移动平台的移动速度数据,检测所述地面可移动平台的运动信息;
其中,所述移动速度数据通过搭载于所述地面可移动平台的轮速采集装置获取。
本发明实施例中,检测模块73,具体用于:
利用所述地面可移动平台的平面特征约束,处理所述移动速度数据,以检测所述地面可移动平台的运动信息。
本发明实施例中,所述平面特征约束用于约束所述地面可移动平台在水平方向上的实际位姿。
具体的,检测模块73,具体用于:
根据所述移动速度数据,获取所述地面可移动平台的位姿数据;
对所述位姿数据进行第一坐标转换处理,得到惯性坐标系下的惯性位姿数据;
获取所述惯性位姿数据的水平分量,得到所述惯性坐标系下的水平位姿 数据;
利用所述平面特征约束处理所述水平位姿数据,以检测所述地面可移动平台的运动信息;
其中,所述平面特征约束包含第一关系,所述第一关系为所述水平位姿数据与所述实际位姿之间的关系。
此外,另一种实现方式中,检测模块73,还用于:
利用所述地面可移动平台的轮速约束,处理所述移动速度数据,以检测所述地面可移动平台的运动信息。
此时,进一步的,检测模块73,具体用于:
对所述移动速度数据进行第二坐标转换处理,得到世界坐标系下的转换速度数据;
利用所述轮速约束处理所述转换速度数据,以检测所述地面可移动平台的运动信息;
其中,所述轮速约束包含第二关系,所述第二关系为所述转换速度数据与所述地面可移动平台的实际速度之间的关系。
此外,另一种实现方式中,检测模块73,还用于:
根据所述移动速度数据,检测所述地面可移动平台是否处于静止状态;
若所述地面可移动平台处于所述静止状态,利用所述地面可移动平台的静止约束处理所述惯性导航数据与所述移动速度数据,以检测所述地面可移动平台的运动信息。
此时,进一步的,检测模块73,具体用于:
根据所述惯性导航数据与所述移动速度数据,获取所述地面可移动平台在初始处于静止状态时的初始位姿数据与初始速度数据;
利用所述静止约束处理所述初始位姿数据与初始速度数据,以检测所述地面可移动平台的运动信息;
其中,所述静止约束包括第三关系与第四关系;其中,所述第三关系为所述初始位姿数据与所述地面可移动平台的实际位姿之间的关系,所述第四关系为所述初始速度数据与所述地面可移动平台的实际速度之间的关系。
此外,该地面可移动平台运动信息检测装置700还可以包括:
第三获取模块(图7未示出),用于获取所述轮速采集装置采集到的至 少一个轮子的轮速数据或转角数据,以作为所述移动速度数据。
本发明实施例中,所述轮速采集装置包括如下至少一种:轮速传感器、轮子编码器。
具体的,所述至少一个轮子包括:所述地面可移动平台的两个后轮。
另一种可能的设计中,检测模块73,具体用于:
利用所述地面可移动平台的环境特征约束,处理所述图像数据,以检测所述地面可移动平台的运动信息;
其中,所述环境特征约束用于约束所述地面可移动平台所在环境。
具体的,检测模块73,具体用于:
对所述至少两帧图像数据进行特征点匹配,得到环境特征点;
利用所述环境特征约束处理所述环境特征点,以检测所述地面可移动平台的运动信息。
另一种可能的设计中,检测模块73,具体用于:
利用所述地面可移动平台的惯性特征约束,处理所述惯性导航数据,以检测所述地面可移动平台的运动信息;
其中,所述惯性特征约束用于约束所述地面可移动平台的位置和姿态。
具体的,检测模块73,具体用于:
根据所述惯性导航数据,获取所述地面可移动平台的速度数据和/或位移数据;
利用所述地面可移动平台的惯性特征约束,处理所述速度数据和/或所述位移数据,以检测所述地面可移动平台的运动信息。
另一种可能的设计中,检测模块73,具体用于:
在所述至少两帧图像数据中确定至少两个关键帧;
根据所述至少两个关键帧之间的所述惯性导航数据,获取所述地面可移动平台的速度数据和/或位移数据;
利用所述地面可移动平台的惯性特征约束,处理所述速度数据和/或所述位移数据,以检测所述地面可移动平台的运动信息。
另一种可能的设计中,第二获取模块72,具体用于:
获取惯性传感器采集得到的所述地面可移动平台的加速度与角速度数据中的至少一种,以作为所述惯性导航数据。
本发明实施例中,所述地面可移动平台包括:车辆。
图7所示实施例的地面可移动平台运动信息检测装置700可用于执行上述方法实施例的技术方案,其实现原理和技术效果可以进一步参考方法实施例中的相关描述,可选的,该地面可移动平台运动信息检测装置700可以为地面可移动平台中的控制器或处理器。
应理解以上图7所示地面可移动平台运动信息检测装置700的各个模块的划分仅仅是一种逻辑功能的划分,实际实现时可以全部或部分集成到一个物理实体上,也可以物理上分开。且这些模块可以全部以软件通过处理元件调用的形式实现;也可以全部以硬件的形式实现;还可以部分模块以软件通过处理元件调用的形式实现,部分模块通过硬件的形式实现。例如,检测模块73可以为单独设立的处理元件,也可以集成在地面可移动平台运动信息检测装置700中,例如终端的某一个芯片中实现,此外,也可以以程序的形式存储于地面可移动平台运动信息检测装置700的存储器中,由地面可移动平台运动信息检测装置700的某一个处理元件调用并执行以上各个模块的功能。其它模块的实现与之类似。此外这些模块全部或部分可以集成在一起,也可以独立实现。这里所述的处理元件可以是一种集成电路,具有信号的处理能力。在实现过程中,上述方法的各步骤或以上各个模块可以通过处理器元件中的硬件的集成逻辑电路或者软件形式的指令完成。
例如,以上这些模块可以是被配置成实施以上方法的一个或多个集成电路,例如:一个或多个特定集成电路(Application Specific Integrated Circuit,ASIC),或,一个或多个微处理器(digital singnal processor,DSP),或,一个或者多个现场可编程门阵列(Field Programmable Gate Array,FPGA)等。再如,当以上某个模块通过处理元件调度程序的形式实现时,该处理元件可以是通用处理器,例如中央处理器(Central Processing Unit,CPU)或其它可以调用程序的处理器。再如,这些模块可以集成在一起,以片上系统(system-on-a-chip,SOC)的形式实现。
并且,本发明实施例提供了一种地面可移动平台运动信息检测装置,请参考图8,该地面可移动平台运动信息检测装置700,包括:
存储器710;
处理器720;以及
计算机程序;
其中,计算机程序存储在存储器710中,并被配置为由处理器720执行以实现如上述实施例所述的方法。
其中,地面可移动平台运动信息检测装置700中处理器720的数目可以为一个或多个,处理器720也可以称为处理单元,可以实现一定的控制功能。所述处理器720可以是通用处理器或者专用处理器等。在一种可选地设计中,处理器720也可以存有指令,所述指令可以被所述处理器720运行,使得所述地面可移动平台运动信息检测装置700执行上述方法实施例中描述的方法。
在又一种可能的设计中,地面可移动平台运动信息检测装置700可以包括电路,所述电路可以实现前述方法实施例中发送或接收或者通信的功能。
可选地,所述地面可移动平台运动信息检测装置700中存储器710的数目可以为一个或多个,存储器710上存有指令或者中间数据,所述指令可在所述处理器720上被运行,使得所述地面可移动平台运动信息检测装置700执行上述方法实施例中描述的方法。可选地,所述存储器710中还可以存储有其他相关数据。可选地处理器720中也可以存储指令和/或数据。所述处理器720和存储器710可以单独设置,也可以集成在一起。
此外,如图8所示,在该地面可移动平台运动信息检测装置700中还设置有收发器730,其中,所述收发器730可以称为收发单元、收发机、收发电路、或者收发器等,用于与测试设备或其他终端设备进行数据传输或通信,在此不再赘述。
如图8所示,存储器710、处理器720与收发器730通过总线连接并通信。
若该地面可移动平台运动信息检测装置700用于实现对应于图2中的方法时,例如,可以由收发器730向各测试终端发布被测包体,收发器730还可以用于接收各测试终端反馈的测试运行数据。而处理器720用于完成相应的确定或者控制操作,可选的,还可以在存储器710中存储相应的指令。各个部件的具体的处理方式可以参考前述实施例的相关描述。
此外,需要说明的是,如图7或图8所示的地面可移动平台运动信息检测装置700可以为一个单独的设备,例如,可以为车辆的辅助驾驶装置(或自动驾驶装置)中的一个可单独安装的处理器或处理装置;或者,该地面可移动平台运动信息检测装置700可以与地面可移动平台中的其他装置进行集 成,例如,可以集成设置在超算平台或存储系统中。
当地面可移动平台运动信息检测装置700为一个单独设置的设备时,地面可移动平台运动信息检测装置700可以仅包含处理器,以实现如实施例一任一实现方式所述的犯犯法,或者,除此之外,还可以进一步包含视觉传感器等设备。而该地面可移动平台运动信息检测装置700一般可直接调用地面可移动平台中搭载的轮速采集装置、惯性传感器、摄像头等采集到的数据。在具体实现时,地面可移动平台运动信息检测装置700可以通过总线或无线通信的方式与地面可移动平台中搭载的其他数据采集设备进行通信,从而获取到这些数据采集设备采集到的各类数据。
此外,本发明实施例提供了一种可读存储介质,其上存储有计算机程序,该计算机程序被处理器执行以实现如实施例一所述的方法。
以及,本发明实施例提供了一种地面可移动平台运动信息检测系统,请参考图9,该地面可移动平台运动信息检测系统900包括:
图像采集装置910,用于获取至少两帧所述图像数据;
惯性传感器920,用于获取所述地面可移动平台的所述惯性导航数据;
轮速采集装置930,用于获取所述地面可移动平台的移动速度数据;
地面可移动平台运动信息检测装置700,用于:
获取所述图像采集装置910获取到的至少两帧所述图像数据,所述图像数据包括所述地面可移动平台周围环境的环境图像数据;
获取所述惯性传感器920获取到的所述惯性导航数据;
根据所述图像数据、所述惯性导航数据与所述地面可移动平台的移动速度数据,检测所述地面可移动平台的运动信息;
其中,所述移动速度数据通过搭载于所述地面可移动平台的所述轮速采集装置930获取。
一种可能的设计中,所述地面可移动平台运动信息检测装置700,具体用于:
利用所述地面可移动平台的平面特征约束,处理所述移动速度数据,以检测所述地面可移动平台的运动信息。
本发明实施例中,所述平面特征约束用于约束所述地面可移动平台在水平方向上的实际位姿。
另一种可能的设计中,所述地面可移动平台运动信息检测装置700,具体用于:
根据所述移动速度数据,获取所述地面可移动平台的位姿数据;
对所述位姿数据进行第一坐标转换处理,得到惯性坐标系下的惯性位姿数据;
获取所述惯性位姿数据的水平分量,得到所述惯性坐标系下的水平位姿数据;
利用所述平面特征约束处理所述水平位姿数据,以检测所述地面可移动平台的运动信息;
其中,所述平面特征约束包含第一关系,所述第一关系为所述水平位姿数据与所述实际位姿之间的关系。
另一种可能的设计中,所述地面可移动平台运动信息检测装置700,还用于:
利用所述地面可移动平台的轮速约束,处理所述移动速度数据,以检测所述地面可移动平台的运动信息。
另一种可能的设计中,所述地面可移动平台运动信息检测装置700,具体用于:
对所述移动速度数据进行第二坐标转换处理,得到世界坐标系下的转换速度数据;
利用所述轮速约束处理所述转换速度数据,以检测所述地面可移动平台的运动信息;
其中,所述轮速约束包含第二关系,所述第二关系为所述转换速度数据与所述地面可移动平台的实际速度之间的关系。
另一种可能的设计中,所述地面可移动平台运动信息检测装置700,还用于:
根据所述移动速度数据,检测所述地面可移动平台是否处于静止状态;
若所述地面可移动平台处于所述静止状态利用所述地面可移动平台的静止约束处理所述惯性导航数据与所述移动速度数据,以检测所述地面可移动平台的运动信息。
另一种可能的设计中,所述地面可移动平台运动信息检测装置700,具 体用于:
根据所述惯性导航数据与所述移动速度数据,获取所述地面可移动平台在初始处于静止状态时的初始位姿数据与初始速度数据;
利用所述静止约束处理所述初始位姿数据与初始速度数据,以检测所述地面可移动平台的运动信息;
其中,所述静止约束包括第三关系与第四关系;其中,所述第三关系为所述初始位姿数据与所述地面可移动平台的实际位姿之间的关系,所述第四关系为所述初始速度数据与所述地面可移动平台的实际速度之间的关系。
另一种可能的设计中,所述地面可移动平台运动信息检测装置700,还用于:
获取所述轮速采集装置930采集到的至少一个轮子的轮速数据或转角数据,以作为所述移动速度数据。
另一种可能的设计中,所述轮速采集装置930包括如下至少一种:轮速传感器、轮子编码器。
另一种可能的设计中,所述至少一个轮子包括:所述地面可移动平台的两个后轮。
另一种可能的设计中,所述地面可移动平台运动信息检测装置700,具体用于:
利用所述地面可移动平台的环境特征约束,处理所述图像数据,以检测所述地面可移动平台的运动信息;
其中,所述环境特征约束用于约束所述地面可移动平台所在环境。
另一种可能的设计中,所述地面可移动平台运动信息检测装置700,具体用于:
对所述至少两帧图像数据进行特征点匹配,得到环境特征点;
利用所述环境特征约束处理所述环境特征点,以检测所述地面可移动平台的运动信息。
另一种可能的设计中,所述地面可移动平台运动信息检测装置700,具体用于:
利用所述地面可移动平台的惯性特征约束,处理所述惯性导航数据,以检测所述地面可移动平台的运动信息;
其中,所述惯性特征约束用于约束所述地面可移动平台的位置和姿态。
另一种可能的设计中,所述地面可移动平台运动信息检测装置700,具体用于:
利用所述地面可移动平台的惯性特征约束,处理所述速度数据和/或所述位移数据,以检测所述地面可移动平台的运动信息。
另一种可能的设计中,所述地面可移动平台运动信息检测装置700,具体用于:
在所述至少两帧图像数据中确定至少两个关键帧;
根据所述至少两个关键帧之间的所述惯性导航数据,获取所述地面可移动平台的速度数据和/或位移数据;
利用所述地面可移动平台的惯性特征约束,处理所述速度数据和/或所述位移数据,以检测所述地面可移动平台的运动信息。
另一种可能的设计中,所述地面可移动平台运动信息检测装置700,具体用于:
获取所述惯性传感器920采集得到的所述地面可移动平台的加速度与角速度数据中的至少一种,以作为所述惯性导航数据。
另一种可能的设计中,所述地面可移动平台包括:车辆。
其中,地面可移动平台运动信息检测装置700、图像采集装置910、惯性传感器920与轮速采集装置930之间,可以通过有线或无线通信的方式进行数据交互与传输。
在一种具体的实现中,该地面可移动平台运动信息检测系统900可以具体为车辆中搭载的ADAS系统。
以及,本发明实施例提供了一种地面可移动平台,请参考图10,该地面可移动平台包括:
机身1010;
动力系统1020,安装于所述机身1010,用于提供运行动力;
地面可移动平台运动信息检测装置700,用于执行如实施例一种任一实现方式所述的方法;
图像采集装置910,用于获取至少两帧所述图像数据;
惯性传感器920,用于获取所述地面可移动平台的所述惯性导航数据;
轮速采集装置930,用于获取所述地面可移动平台的移动速度数据。
本发明实施例对于地面可移动平台中机身1010和动力系统1020的具体结构及组成方式无限定,在此不做赘述。
如前所述,本发明实施例所涉及到的地面可移动平台,可以包括但不限于:车辆。除此之外,还可以包括其他地面机器装置,其中,地面机器装置可以包括但不限于:地面机器玩具与地面机器人中的至少一种。例如,地面机器玩具可以为玩具汽车,又例如,地面机器人可以为地面扫地机器人、地面调度机器人等。
由于本实施例中的各模块能够执行实施例一所示的方法,本实施例未详细描述的部分,可参考对实施例一的相关说明。
本领域普通技术人员可以理解:实现上述方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成,前述的程序可以存储于一计算机可读取存储介质中,该程序在执行时,执行包括上述方法实施例的步骤;而前述的存储介质包括:ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是:以上各实施例仅用以说明本发明的技术方案,而非对其限制;尽管参照前述各实施例对本发明进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分或者全部技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案的本质脱离本发明各实施例技术方案的范围。
Claims (37)
- 一种地面可移动平台运动信息检测方法,其特征在于,包括:获取至少两帧图像数据,所述图像数据包括所述地面可移动平台周围环境的环境图像数据;获取所述地面可移动平台的惯性导航数据;根据所述图像数据、所述惯性导航数据与所述地面可移动平台的移动速度数据,检测所述地面可移动平台的运动信息;其中,所述移动速度数据通过搭载于所述地面可移动平台的轮速采集装置获取。
- 根据权利要求1所述的方法,其特征在于,根据所述地面可移动平台的移动速度数据,检测所述地面可移动平台的运动信息,包括:利用所述地面可移动平台的平面特征约束,处理所述移动速度数据,以检测所述地面可移动平台的运动信息。
- 根据权利要求2所述的方法,其特征在于,所述平面特征约束用于约束所述地面可移动平台在水平方向上的实际位姿。
- 根据权利要求2所述的方法,其特征在于,所述利用所述地面可移动平台的平面特征约束,处理所述移动速度数据,包括:根据所述移动速度数据,获取所述地面可移动平台的位姿数据;对所述位姿数据进行第一坐标转换处理,得到惯性坐标系下的惯性位姿数据;获取所述惯性位姿数据的水平分量,得到所述惯性坐标系下的水平位姿数据;利用所述平面特征约束处理所述水平位姿数据,以检测所述地面可移动平台的运动信息;其中,所述平面特征约束包含第一关系,所述第一关系为所述水平位姿数据与所述地面可移动平台的实际位姿之间的关系。
- 根据权利要求1-4任一项所述的方法,其特征在于,所述方法还包括:利用所述地面可移动平台的轮速约束,处理所述移动速度数据,以检测所述地面可移动平台的运动信息。
- 根据权利要求5所述的方法,其特征在于,所述利用所述地面可移动 平台的轮速约束,处理所述移动速度数据,包括:对所述移动速度数据进行第二坐标转换处理,得到世界坐标系下的转换速度数据;利用所述轮速约束处理所述转换速度数据,以检测所述地面可移动平台的运动信息;其中,所述轮速约束包含第二关系,所述第二关系为所述转换速度数据与所述地面可移动平台的实际速度之间的关系。
- 根据权利要求1-4任一项所述的方法,其特征在于,所述方法还包括:根据所述移动速度数据,检测所述地面可移动平台是否处于静止状态;若所述地面可移动平台处于所述静止状态,利用所述地面可移动平台的静止约束处理所述惯性导航数据与所述移动速度数据,以检测所述地面可移动平台的运动信息。
- 根据权利要求7所述的方法,其特征在于,所述利用所述地面可移动平台的静止约束处理所述惯性导航数据与所述移动速度数据,包括:根据所述惯性导航数据与所述移动速度数据,获取所述地面可移动平台在初始处于静止状态时的初始位姿数据与初始速度数据;利用所述静止约束处理所述初始位姿数据与初始速度数据,以检测所述地面可移动平台的运动信息;其中,所述静止约束包括第三关系与第四关系;其中,所述第三关系为所述初始位姿数据与所述地面可移动平台的实际位姿之间的关系,所述第四关系为所述初始速度数据与所述地面可移动平台的实际速度之间的关系。
- 根据权利要求1-4任一项所述的方法,其特征在于,所述方法还包括:获取所述轮速采集装置采集到的至少一个轮子的轮速数据或转角数据,以作为所述移动速度数据。
- 根据权利要求9所述的方法,其特征在于,所述轮速采集装置包括如下至少一种:轮速传感器、轮子编码器。
- 根据权利要求9所述的方法,其特征在于,所述至少一个轮子包括:所述地面可移动平台的两个后轮。
- 根据权利要求1所述的方法,其特征在于,根据所述图像数据,检测所述地面可移动平台的运动信息,包括:利用所述地面可移动平台的环境特征约束,处理所述图像数据,以检测所述地面可移动平台的运动信息;其中,所述环境特征约束用于约束所述地面可移动平台所在环境。
- 根据权利要求12所述的方法,其特征在于,所述利用所述地面可移动平台的环境特征约束,处理所述图像数据,以检测所述地面可移动平台的运动信息,包括:对所述至少两帧图像数据进行特征点匹配,得到环境特征点;利用所述环境特征约束处理所述环境特征点,以检测所述地面可移动平台的运动信息。
- 根据权利要求1所述的方法,其特征在于,根据所述惯性导航数据,检测所述地面可移动平台的运动信息,包括:利用所述地面可移动平台的惯性特征约束,处理所述惯性导航数据,以检测所述地面可移动平台的运动信息;其中,所述惯性特征约束用于约束所述地面可移动平台的位置和姿态。
- 根据权利要求14所述的方法,其特征在于,所述利用所述地面可移动平台的惯性特征约束,处理所述惯性导航数据,以检测所述地面可移动平台的运动信息,包括:根据所述惯性导航数据,获取所述地面可移动平台的速度数据和/或位移数据;利用所述地面可移动平台的惯性特征约束,处理所述速度数据和/或所述位移数据,以检测所述地面可移动平台的运动信息。
- 根据权利要求14所述的方法,其特征在于,所述利用所述地面可移动平台的惯性特征约束,处理所述惯性导航数据,以检测所述地面可移动平台的运动信息,包括:在所述至少两帧图像数据中确定至少两个关键帧;根据所述至少两个关键帧之间的所述惯性导航数据,获取所述地面可移动平台的速度数据和/或位移数据;利用所述地面可移动平台的惯性特征约束,处理所述速度数据和/或所述位移数据,以检测所述地面可移动平台的运动信息。
- 根据权利要求1-4、12-16中任一项所述的方法,其特征在于,所述 获取所述地面可移动平台的惯性导航数据,包括:获取惯性传感器采集得到的所述地面可移动平台的加速度与角速度数据中的至少一种,以作为所述惯性导航数据。
- 根据权利要求1-4、12-16中任一项所述的方法,其特征在于,所述地面可移动平台包括:车辆。
- 一种地面可移动平台运动信息检测系统,其特征在于,包括:图像采集装置,用于获取至少两帧所述图像数据;惯性传感器,用于获取所述地面可移动平台的所述惯性导航数据;轮速采集装置,用于获取所述地面可移动平台的移动速度数据;地面可移动平台运动信息检测装置,用于:获取所述图像采集装置获取到的至少两帧所述图像数据,所述图像数据包括所述地面可移动平台周围环境的环境图像数据;获取所述惯性传感器获取到的所述惯性导航数据;根据所述图像数据、所述惯性导航数据与所述地面可移动平台的移动速度数据,检测所述地面可移动平台的运动信息;其中,所述移动速度数据通过搭载于所述地面可移动平台的所述轮速采集装置获取。
- 根据权利要求19所述的系统,其特征在于,所述地面可移动平台运动信息检测装置,具体用于:利用所述地面可移动平台的平面特征约束,处理所述移动速度数据,以检测所述地面可移动平台的运动信息。
- 根据权利要求20所述的系统,其特征在于,所述平面特征约束用于约束所述地面可移动平台在水平方向上的实际位姿。
- 根据权利要求20所述的系统,其特征在于,所述地面可移动平台运动信息检测装置,具体用于:根据所述移动速度数据,获取所述地面可移动平台的位姿数据;对所述位姿数据进行第一坐标转换处理,得到惯性坐标系下的惯性位姿数据;获取所述惯性位姿数据的水平分量,得到所述惯性坐标系下的水平位姿 数据;利用所述平面特征约束处理所述水平位姿数据,以检测所述地面可移动平台的运动信息;其中,所述平面特征约束包含第一关系,所述第一关系为所述水平位姿数据与所述地面可移动平台的实际位姿之间的关系。
- 根据权利要求19-22任一项所述的系统,其特征在于,所述地面可移动平台运动信息检测装置,还用于:利用所述地面可移动平台的轮速约束,处理所述移动速度数据,以检测所述地面可移动平台的运动信息。
- 根据权利要求23所述的系统,其特征在于,所述地面可移动平台运动信息检测装置,具体用于:对所述移动速度数据进行第二坐标转换处理,得到世界坐标系下的转换速度数据;利用所述轮速约束处理所述转换速度数据,以检测所述地面可移动平台的运动信息;其中,所述轮速约束包含第二关系,所述第二关系为所述转换速度数据与所述地面可移动平台的实际速度之间的关系。
- 根据权利要求19-22任一项所述的系统,其特征在于,所述地面可移动平台运动信息检测装置,还用于:根据所述移动速度数据,检测所述地面可移动平台是否处于静止状态;若所述地面可移动平台处于所述静止状态,利用所述地面可移动平台的静止约束处理所述惯性导航数据与所述移动速度数据,以检测所述地面可移动平台的运动信息。
- 根据权利要求25所述的系统,其特征在于,所述地面可移动平台运动信息检测装置,具体用于:根据所述惯性导航数据与所述移动速度数据,获取所述地面可移动平台在初始处于静止状态时的初始位姿数据与初始速度数据;利用所述静止约束处理所述初始位姿数据与初始速度数据,以检测所述地面可移动平台的运动信息;其中,所述静止约束包括第三关系与第四关系;其中,所述第三关系为 所述初始位姿数据与所述地面可移动平台的实际位姿之间的关系,所述第四关系为所述初始速度数据与所述地面可移动平台的实际速度之间的关系。
- 根据权利要求19-22任一项所述的系统,其特征在于,所述地面可移动平台运动信息检测装置,还用于:获取所述轮速采集装置采集到的至少一个轮子的轮速数据或转角数据,以作为所述移动速度数据。
- 根据权利要求27所述的系统,其特征在于,所述轮速采集装置包括如下至少一种:轮速传感器、轮子编码器。
- 根据权利要求27所述的系统,其特征在于,所述至少一个轮子包括:所述地面可移动平台的两个后轮。
- 根据权利要求19所述的系统,其特征在于,所述地面可移动平台运动信息检测装置,具体用于:利用所述地面可移动平台的环境特征约束,处理所述图像数据,以检测所述地面可移动平台的运动信息;其中,所述环境特征约束用于约束所述地面可移动平台所在环境。
- 根据权利要求30所述的系统,其特征在于,所述地面可移动平台运动信息检测装置,具体用于:对所述至少两帧图像数据进行特征点匹配,得到环境特征点;利用所述环境特征约束处理所述环境特征点,以检测所述地面可移动平台的运动信息。
- 根据权利要求19所述的系统,其特征在于,所述地面可移动平台运动信息检测装置,具体用于:利用所述地面可移动平台的惯性特征约束,处理所述惯性导航数据,以检测所述地面可移动平台的运动信息;其中,所述惯性特征约束用于约束所述地面可移动平台的位置和姿态。
- 根据权利要求32所述的系统,其特征在于,所述地面可移动平台运动信息检测装置,具体用于:根据所述惯性导航数据,获取所述地面可移动平台的速度数据和/或位移数据;利用所述地面可移动平台的惯性特征约束,处理所述速度数据和/或所述 位移数据,以检测所述地面可移动平台的运动信息。
- 根据权利要求32所述的系统,其特征在于,所述地面可移动平台运动信息检测装置,具体用于:在所述至少两帧图像数据中确定至少两个关键帧;根据所述至少两个关键帧之间的所述惯性导航数据,获取所述地面可移动平台的速度数据和/或位移数据;利用所述地面可移动平台的惯性特征约束,处理所述速度数据和/或所述位移数据,以检测所述地面可移动平台的运动信息。
- 根据权利要求19-22、30-34中任一项所述的系统,其特征在于,所述地面可移动平台运动信息检测装置,具体用于:获取所述惯性传感器采集得到的所述地面可移动平台的加速度与角速度数据中的至少一种,以作为所述惯性导航数据。
- 根据权利要求19-22、30-34中任一项所述的系统,其特征在于,所述地面可移动平台包括:车辆。
- 一种地面可移动平台,其特征在于,包括:机身;动力系统,安装于所述机身,用于提供运行动力;地面可移动平台运动信息检测装置,用于执行如权利要求1至16任一项所述的方法;图像采集装置,用于获取至少两帧所述图像数据;惯性传感器,用于获取所述地面可移动平台的所述惯性导航数据;轮速采集装置,用于获取所述地面可移动平台的移动速度数据。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201980011862.5A CN111699363A (zh) | 2019-05-28 | 2019-05-28 | 地面可移动平台与其运动信息检测方法、系统 |
PCT/CN2019/088757 WO2020237495A1 (zh) | 2019-05-28 | 2019-05-28 | 地面可移动平台与其运动信息检测方法、系统 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2019/088757 WO2020237495A1 (zh) | 2019-05-28 | 2019-05-28 | 地面可移动平台与其运动信息检测方法、系统 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020237495A1 true WO2020237495A1 (zh) | 2020-12-03 |
Family
ID=72476449
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2019/088757 WO2020237495A1 (zh) | 2019-05-28 | 2019-05-28 | 地面可移动平台与其运动信息检测方法、系统 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN111699363A (zh) |
WO (1) | WO2020237495A1 (zh) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115575923A (zh) * | 2022-12-08 | 2023-01-06 | 千巡科技(深圳)有限公司 | 一种地面机器人静止判断方法、系统、装置以及存储介质 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012255686A (ja) * | 2011-06-08 | 2012-12-27 | Mitsubishi Electric Corp | 車両の位置姿勢測定装置 |
EP2930467A1 (en) * | 2014-04-11 | 2015-10-14 | Airbus Defence and Space GmbH | A system and method for sensing the inclination of a moving platform with respect to gravity |
CN107463173A (zh) * | 2017-07-31 | 2017-12-12 | 广州维绅科技有限公司 | 仓储agv导航方法及装置、计算机设备及存储介质 |
CN207881711U (zh) * | 2018-02-05 | 2018-09-18 | 深圳鼎然信息科技有限公司 | 基于gnss的惯性导航系统 |
CN108731667A (zh) * | 2017-04-14 | 2018-11-02 | 百度在线网络技术(北京)有限公司 | 用于确定无人驾驶车辆的速度和位姿的方法和装置 |
CN109579844A (zh) * | 2018-12-04 | 2019-04-05 | 电子科技大学 | 定位方法及系统 |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102435188B (zh) * | 2011-09-15 | 2013-10-02 | 南京航空航天大学 | 一种用于室内环境的单目视觉/惯性全自主导航方法 |
CN107255476B (zh) * | 2017-07-06 | 2020-04-21 | 青岛海通胜行智能科技有限公司 | 一种基于惯性数据和视觉特征的室内定位方法和装置 |
-
2019
- 2019-05-28 WO PCT/CN2019/088757 patent/WO2020237495A1/zh active Application Filing
- 2019-05-28 CN CN201980011862.5A patent/CN111699363A/zh active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012255686A (ja) * | 2011-06-08 | 2012-12-27 | Mitsubishi Electric Corp | 車両の位置姿勢測定装置 |
EP2930467A1 (en) * | 2014-04-11 | 2015-10-14 | Airbus Defence and Space GmbH | A system and method for sensing the inclination of a moving platform with respect to gravity |
CN108731667A (zh) * | 2017-04-14 | 2018-11-02 | 百度在线网络技术(北京)有限公司 | 用于确定无人驾驶车辆的速度和位姿的方法和装置 |
CN107463173A (zh) * | 2017-07-31 | 2017-12-12 | 广州维绅科技有限公司 | 仓储agv导航方法及装置、计算机设备及存储介质 |
CN207881711U (zh) * | 2018-02-05 | 2018-09-18 | 深圳鼎然信息科技有限公司 | 基于gnss的惯性导航系统 |
CN109579844A (zh) * | 2018-12-04 | 2019-04-05 | 电子科技大学 | 定位方法及系统 |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115575923A (zh) * | 2022-12-08 | 2023-01-06 | 千巡科技(深圳)有限公司 | 一种地面机器人静止判断方法、系统、装置以及存储介质 |
Also Published As
Publication number | Publication date |
---|---|
CN111699363A (zh) | 2020-09-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP7073315B2 (ja) | 乗物、乗物測位システム、及び乗物測位方法 | |
CN110009681B (zh) | 一种基于imu辅助的单目视觉里程计位姿处理方法 | |
CN111595333A (zh) | 视觉惯性激光数据融合的模块化无人车定位方法及系统 | |
JP6141601B2 (ja) | 車載カメラ自動キャリブレーション装置 | |
US20170017839A1 (en) | Object detection apparatus, object detection method, and mobile robot | |
JP2019049457A (ja) | 画像処理装置および測距装置 | |
CN110411457B (zh) | 基于行程感知与视觉融合的定位方法、系统、终端和存储介质 | |
CN106056664A (zh) | 一种基于惯性和深度视觉的实时三维场景重构系统及方法 | |
CN112669354B (zh) | 一种基于车辆非完整约束的多相机运动状态估计方法 | |
CN108489454A (zh) | 深度距离测量方法、装置、计算机可读存储介质及电子设备 | |
CN112017236B (zh) | 一种基于单目相机计算目标物位置的方法及装置 | |
CN110207693B (zh) | 一种鲁棒立体视觉惯性预积分slam方法 | |
WO2017057052A1 (ja) | 情報処理装置、情報処理方法、およびプログラム | |
CN111890373A (zh) | 车载机械臂的感知定位方法 | |
CN116443028A (zh) | 头部姿态数据采集系统及方法 | |
WO2020237495A1 (zh) | 地面可移动平台与其运动信息检测方法、系统 | |
CN112819898A (zh) | 热像仪-rgb相机-imu空间联合标定方法、系统和存储介质 | |
CN116182855B (zh) | 一种弱光强环境下仿复眼偏振视觉无人机组合导航方法 | |
CN116952229A (zh) | 无人机定位方法、装置、系统和存储介质 | |
CN113252066A (zh) | 里程计设备参数的标定方法及装置、存储介质、电子装置 | |
US11740103B2 (en) | Map creation device, map creation system, map creation method, and storage medium | |
CN116665166A (zh) | 一种适用于不平坦路面场景的智能车辆3d目标检测方法 | |
CN110309845A (zh) | 信息处理系统以及信息处理方法 | |
CN116659490A (zh) | 低成本视觉-惯性融合的slam方法 | |
CN116804553A (zh) | 基于事件相机/imu/自然路标的里程计系统及方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19930908 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19930908 Country of ref document: EP Kind code of ref document: A1 |