JPH09264955A - Obstacle sensing apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to rapidly, accurately process obstacle data by predicting the detecting time to be detected next for preceding vehicle detected this time, and unifying both preceding vehicles based on the comparison of predicting detecting time with real detecting time. SOLUTION: A laser head unit 1 oscillates a laser beam from an oscillator 2a toward the forward of a vehicle, receives the reflected wave by a receiver 1b, and detects the preceding vehicle. In this case, the oscillating direction is altered by an oscillating direction altering unit 1c, and the angle and distance of an object are detected at each altering timing. A time predicting unit 2 calculates the predicting detecting time of the preceding vehicle detected by the unit 1 this time, and a position predicting unit 3 calculates the predicting detecting position of the preceding vehicle at the predicted detecting time calculated by the unit 2. An unification discriminator 4 compares the predicted and real detected times calculated by the units 2, 3 with the predicted and real detected positions, and discriminates whether the preceding vehicles detected this time and next time are the same or not.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention detects an obstacle by detecting an object such as a preceding vehicle traveling in front of an automobile by a reflected wave of a detection wave whose transmitting direction changes at a predetermined angular velocity within a predetermined range. The present invention relates to an obstacle detection device, and more particularly, to a measure for reducing a processing load required to make an object detected at each transmission timing of a detected wave uniform.

[0002]

2. Description of the Related Art An obstacle detecting device of this type is known, for example, from Japanese Patent Application Laid-Open No. 3-111785. In this model, the horizontal dimension of the object is calculated by changing the transmission direction so that the transmission direction of the radar beam reciprocates at a predetermined angular velocity within a predetermined range that spreads horizontally in front of the host vehicle. By making it possible to recognize an object as a preceding vehicle based on the CPU, the number of objects to be detected as obstacles is reduced, and a CPU required for obstacle detection is
The calculation processing time can be shortened.

When executing ICC (Intelligent Auto Cruise) or CW (Collision Warning), if there are a plurality of preceding vehicles ahead of the host vehicle, the same timing is used at each timing of changing the transmission direction. How to identify the preceding vehicle is a big problem. For example, for a preceding vehicle shown by a white triangle in FIG. 12, as shown in FIG. 12A, this time, the angular position R ₁ detected while the transmission direction changes from the left side to the right side in the figure in a fan shape. And the detection time t _1, and as shown in FIG.
Next time, when the relative angular velocity V is to be obtained based on the angular position R ₈ of the preceding vehicle and the detection time t ₈ detected while the transmission direction changes from the right side to the left side in the case of a human being, If so, the following processing is performed.

V = (R ₈ −R ₁ ) / (t ₈ −t ₁ ) At this time, the human is based on the total judgment of the color and shape of the target visually, and (R ₁ , t ₁ ) and each data that (R _8, t ₈₎ are a recognition that is identical preceding vehicle. However, depending on the radar beam, neither the color nor the shape can be recognized, and generally only the angular position and the detection time are known.

In the conventional obstacle detecting device, how is the relative angular velocity obtained? The next time, with respect to the preceding vehicle of (R ₁ , t ₁ ) detected this time, It is assumed that all the detected preceding vehicles are the same target, and their data (R ₅ , t ₅ ) ~
(R ₈ , t ₈ ) is exhaustively processed by the above equation, and the preceding vehicle having the smallest speed among them, that is, the speed having the smallest moving distance which can be said to be stochastically most likely is regarded as the same target. Judgment (this is defined as identification)
It has been made to be.

[0006]

However, in the processing required to make the targets identical to each other, as the number of detected targets increases, the calculation load increases and the calculation speed of the CPU cannot keep up with the obstacles. It is a problem in processing the data quickly.

As a countermeasure for the above, the position where each preceding vehicle can move is predicted by the next time, and only the preceding vehicle detected next time at the predicted position is subjected to the identification process. It is possible.

However, even in the above measures, when a plurality of preceding vehicles are detected within the predicted position, all the detected preceding vehicles must be exhaustively processed for comparison. The above problems still remain.

Further, since the radar beam emission direction changes at a constant angular velocity, there is a difference in the time detected by each preceding vehicle even at the same measurement timing. Therefore, it is difficult to predict the detected position with sufficient accuracy, and the identification process based on the predicted position with such insufficient accuracy is difficult in accurately processing obstacle data. There is.

The present invention has been made in view of the above points, and its main purpose is to advance a vehicle ahead of a vehicle by a reflected wave of a detection wave whose transmitting direction changes within a predetermined range at a predetermined angular velocity. In an obstacle detection device that detects an object such as a vehicle and detects an obstacle, by adding a new algorithm, an object to be compared at the time of performing the identification processing of the object detected at each change timing of the transmission direction Is to narrow down the processing load so that obstacle data can be processed quickly and accurately.

[0011]

In order to achieve the above object, the present invention predicts the detection time that will be detected next time for the preceding vehicle detected by the radar beam this time, and predicts the detection time. By performing the equalization process on both preceding vehicles based on the comparison between the detection time and the actual detection time, it is possible to narrow down the comparison target at the time of the identification.

Specifically, in the invention of claim 1, as shown in FIG. 1, a transmitter 1a for transmitting a detected wave, a receiver 1b for receiving a reflected wave of the detected wave from an object, and a detector 1b. And a transmission direction changing unit 1c for changing the transmission direction so that the transmission direction of the wave changes within a predetermined range at a predetermined angular velocity. It is premised on an obstacle detection device which detects an obstacle based on the displacement amount and the displacement direction of the object detected by the object detection means 1 within a predetermined time.

Then, this time, the displacement characteristic of the object detected by the object detecting means 1 (displacement amount and displacement direction).
And a predicted detection time calculated by the time predicting means 2 next time, which calculates the predicted detection time of the object detected by the object detecting means 1 next time based on the change characteristic of the transmission direction. Next time, based on the comparison with the actual detection time when the object is detected by the object detecting means 1, the object detected by the object detecting means 1 this time is the same as the object detected next time. The same determination means 4 for determining whether or not there is is provided.

In the above structure, the object detecting means 1 is
By transmitting the detection wave and changing the transmission direction so that the transmission direction changes within a predetermined range at a predetermined angular velocity, an object is detected at each change timing. On the other hand, the predicted detection time of the object that will be detected by the object detection unit 1 next time is calculated by the time prediction unit 2 based on the displacement characteristic of the object and the change characteristic of the transmission direction. Then, when the object is actually detected by the object detecting means 1 next time, the object detected by the object detecting means 1 this time is based on the comparison between the actual detection time and the predicted detection time, The same determination means 4 determines whether or not the object detected in the next scan is the same. As a result, the comparison target of the identification by the identification determination unit 4 is narrowed down to the object detected at the predicted detection time, so that the processing load required for the identification is reduced by the narrowed down comparison target. Obstacle data will be processed quickly and with high accuracy.

According to the invention of claim 2, in the invention of claim 1, the object detecting means 1 will detect the object next time based on the displacement characteristic of the object detected by the object detecting means 1 this time. The position prediction means 3 for calculating the predicted detection position of the object is provided. Then, in addition to comparing the predicted detection time with the actual detection time, the same determination means 4 compares the predicted detection position calculated by the position prediction means 3 with the object detection means 1 to determine whether the object is actually detected next time. Based on both the comparison with the actual detection position at the time of detection, it is determined whether or not the object detected this time by the object detection unit 1 and the object detected next time are the same. It has been done.

In the above configuration, the predicted detection time of the object that will be detected by the object detecting means 1 next time is calculated by the time predicting means 2, while the object detecting means 1 next time.
The predicted detection position of the object that will be detected at is calculated by the position prediction means 3 based on the displacement characteristics of the object. Then, next time, when the object is actually detected at the predicted detection position by the object detection means 1 at the predicted detection time, not only the predicted detection time and the actual detection time are compared, but also the predicted detection position and the actual detection are detected. Also based on the comparison of the positions, the same determination unit 4 determines whether the object detected this time by the object detection unit 1 is the same as the object detected next time. As a result, the comparison target for identification by the identification determination unit 4 is narrowed down to the object detected at the predicted detection time and further narrowed down to the object detected at the predicted detection position.
The processing load required for the identification can be further reduced.

According to the invention of claim 3, on the same premise as in the case of the invention of claim 1, the displacement characteristic of the object detected this time by the object detecting means 1 (displacement amount and direction of displacement).
And a time prediction unit 2 for calculating a predicted detection time of the object that will be detected by the object detection unit 1 next time based on the changing characteristic of the transmission direction of the object detection unit 1.
This time, a position for calculating the predicted detection position of the object that will be detected at the predicted detection time calculated by the time prediction unit 2 based on the displacement characteristic of the object detected by the object detection unit 1. This time, based on the comparison between the prediction means 3, the predicted detection position calculated by the position prediction means 3, and the actual detection position when the object is actually detected by the object detection means 1 next time, Detection means 1
The same determination unit 4 that determines whether or not the object detected by and the object detected next time are the same.

In the above structure, the object detecting means 1 is
By transmitting the detection wave and changing the transmission direction so that the transmission direction changes within a predetermined range at a predetermined angular velocity, an object is detected at each change timing. On the other hand, based on the displacement characteristic of the object and the changing characteristic of the transmitting direction of the object detecting means 1, first, the predicted detection time of the object which will be detected by the object detecting means 1 next time is the time. It is calculated by the prediction means 2. Next, this time, the predicted detection position of the object that will be detected at the predicted detection time calculated by the time prediction unit 2 based on the displacement characteristic of the object detected by the object detection unit 1 is: It is calculated by the position predicting means 3. Then, next time when the object is actually detected by the object detecting means 1, the object detected by the object detecting means 1 this time based on the comparison between the actual detected position and the predicted detected position, The same determination means 4 determines whether or not the object detected next time is the same. That is, since the predicted detection position that will be detected at the next transmission direction change timing is calculated based on the predicted detection time of each object, even if the detection time varies for each object, The detected position is predicted accurately.

According to a fourth aspect of the present invention, in the above-mentioned third aspect, the same determination means 4 compares the predicted detection position with the actual detection position, and the predicted detection time calculated by the time prediction means 2. Next time, based on the comparison with the actual detection time when the object is detected by the object detecting means 1,
It is assumed that it is configured to determine whether the object detected by the object detection means 1 and the object detected the next time are the same.

In the above structure, the comparison target of the identification by the identification determination means 4 is narrowed down to the object at the predicted position calculated based on the predicted detection time, and is detected at the predicted detection time. Since the objects are further narrowed down, the processing load required for identification is further reduced.

According to a fifth aspect of the present invention, in the first to fourth aspects of the invention, based on the predicted detection time calculated by the time predicting means 2, the object detecting means 1 detects the object in a predetermined area next time. The same as the detection determining means 5 for determining whether or not is possible, and the next time the detection determining means 5 determines that the object detecting means 1 cannot detect an object in a predetermined area next time. The management stopping means 6 for excluding the object from the comparison target of the same judgment by the judging means 4 is provided.

In the above structure, the detection determination means 5 is based on the predicted detection time calculated by the time prediction means 2.
Thus, it is determined whether or not the object can be detected by the object detecting means 1 in the predetermined area next time. Then, when the detection determining means 5 determines that the next time the object detecting means 1 cannot detect the object within the predetermined area, the management stopping means 6 compares the identification by the same determining means 4. The object is excluded from the target. By these,
The processing load required for the identification is reduced by the amount of the object determined to be undetectable within the predetermined area at the time of the next detection by the object detection unit 1.

[0023]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 2 shows an automobile equipped with an obstacle detection device according to an embodiment of the present invention. At the front end of the vehicle body of this automobile, an object detecting means for detecting a preceding vehicle as an object existing in front of the automobile. The radar head unit 1 is disposed. The radar head unit 1 emits a laser beam as a detection wave from the transmission unit 1a toward the front of the vehicle, while receiving the reflected wave at the reception unit 1b, thereby detecting the preceding vehicle. Has been done. At this time, the laser beam of the transmitting section 1a is reciprocally changed at a constant angular velocity in a fan shape of a predetermined angle in the horizontal plane by the transmitting direction changing section 1c to change the transmitting direction. Then, the angle and distance of the object are detected at each change timing.

Further, the automobile is equipped with a control unit C to which the detection signal of the radar head unit 1 is input, and the control unit C processes the detection signal. The processing result is, as shown in FIG. 3, a head-up display 12 and an alarm device 1.
3 and the vehicle control device 14 respectively. Then, when an obstacle is detected in front of the host vehicle, the fact is displayed on the front position of the front window together with the state of the traveling road by the head-up display 12, and an alarm is issued by the alarm device 13. On the other hand, the vehicle controller 14 causes each brake 14
.. (only the brake 14a for the left front wheel is shown in FIG. 2) actuate to apply a braking force to each wheel. Further, the control unit C includes a vehicle speed sensor 9 for detecting a vehicle speed of the own vehicle, and a steering angle sensor 1 for detecting a steering angle of a steering handle of the own vehicle.
0 and a yaw rate sensor 11 for detecting the yaw rate generated by the host vehicle are connected to each other.

In the present embodiment, the control unit C is provided with displacement characteristics such as the displacement amount of the preceding vehicle detected by the radar head unit 1 and the direction of the displacement and the transmission direction of the head unit 1 this time. A time predicting unit 2 as a time predicting unit that calculates the predicted detection time of the preceding vehicle that will be detected by the head unit 1 this time based on the change characteristics, and this time is detected by the radar head unit 1. Based on the displacement characteristics of the preceding vehicle, the position predicting unit 3 as a position predicting unit that calculates the predicted detected position of the preceding vehicle that will be detected at the predicted detection time calculated by the time predicting unit 2. And the predicted detection time calculated by the time prediction unit 2 and the actual detection time when actually detected, and the position prediction unit 3 The predicted detected position (angle and distance in this example) calculated by the above is compared with the actual detected position when actually detected, and the preceding vehicle detected this time and the preceding vehicle detected next time are the same. The same determination unit 4 as the same determination means for determining whether or not Also,
The control unit C includes a traveling route estimating unit 7 that estimates the traveling route of the host vehicle based on the detection signals of the sensors 9 to 11, an output signal of the traveling route estimating unit 7, and the processing of the identity determining unit 4. An obstacle determination unit 8 that determines an obstacle based on the signal is also provided.

The traveling path estimating unit 7 estimates the traveling path of the host vehicle based on the vehicle speed V and the steering angle φ of the host vehicle detected by the vehicle speed sensor 9 and the steering angle sensor 10, respectively. Is calculated by calculating the radius of curvature R ₁ of the traveling path using the following equation. _{R 1 = (1 + A ·} V 2) · L B · N / φ where, A: stability factor N: steering gear ratio L _B: wheelbase

The traveling path estimating unit 7 can also estimate the traveling path based on the yaw rate γ of the vehicle and the vehicle speed V detected by the yaw rate sensor 11, and the radius of curvature R of the traveling path at that time can be estimated. ₂ is calculated by the following formula. R ₂ = V / γ

By the way, when there is a cant on a curved road such as a highway, the steering angle φ does not match the actual turning angle of the host vehicle, and the host vehicle estimated based on the steering angle φ. The radius of curvature of the traveling path of is larger than the actual radius of curvature. Further, even when the host vehicle is traveling straight ahead, the steering wheel is generally delicately steered to the left or right. Therefore, if the traveling path is estimated by following such changes in the steering angle φ, The traveled route may not match the actual traveled route.

Therefore, when the steering angle φ is smaller than a predetermined value, the latter radius of curvature R ₂ is selected, and when the steering angle φ is a predetermined value or more, two radiuses of curvature R ₁ ,
It is preferable to select the smaller one of R ₂ . That is, when the vehicle is turning on a curved road with a cant, the vehicle can make a turning motion by the cant without steering the steering wheel to a large extent.
The traveling path is properly estimated by obtaining the radius of curvature R ₂ based on the yaw rate γ generated in the host vehicle. When the host vehicle makes a sharp turn, the radius of curvature R ₁ corresponding to a large steering angle is selected. On the other hand, when the host vehicle runs in a straight line, the yaw rate γ is small even if the steering wheel is slightly operated. Since it hardly occurs, a small radius of curvature R ₂ corresponding to the yaw rate γ is selected.

Here, the predicted detection time (t = t _{m + 1} ) at the current detection time (t = t _m ) in the time prediction unit 2
The calculation method of will be described. In the present embodiment, when the predicted detection position (predicted distance and predicted angle) at the predicted detection time (t _{m + 1} ) is calculated, the predicted detection time (t
_{Since the} predicted detection time interval T ₂ derived from ( _{m + 1} ) is used, the calculation method of the predicted detection time interval T ₂ will also be described.

First, in the case where a preceding vehicle is detected when the laser beam angle θ changes from θ = 0 to θ = θmax in the forward direction, the horizontal axis represents the time t and the vertical axis represents the angle θ. This will be described with reference to FIG. 4 showing the coordinates. In the figure, the change trajectory of the laser beam in the forward direction indicated by # 1, # 2
Reverse direction (direction from θ = θmax to θ = 0)
And the change locus of the angular position of the preceding vehicle as an example indicated by # 3 are as in the following equations (1) to (3). _{θ = θ B · t ... (} 1) θ = -θ B · t + 2 · θmax ... (2) θ = θ T · t + θ k + θ T · (-t m) ... (3) However, point B (t _{m + 1, θ k (t m +} 1)): the predicted detection coordinate position of the preceding vehicle (unknown) point _{a (t m, θ k (} t m)): current detection coordinate position of the preceding vehicle (known) .theta.max: laser beam Angle of change (fixed value) θ _B : Angular velocity of laser beam (fixed value) θ _T : Moving angular velocity of preceding vehicle (unknown)

Solving the predicted detection time (t = t _{m + 1} ) of the point B in this case from the above equations (2) and (3), t _{m + 1} = (θ _T · t _m + 2 · θ max −θ _k ) / (θ _B + θ _T ) ... (4)

Then, the moving angular velocity θ of the preceding vehicle
_As shown in FIG. 5, _T can be obtained from the previous detection time t _m-1 and the previous detection angle θ _{k at} that time, and the current detection time t _m and the current detection angle θ _{k at} that time. In the figure, T ₁ represents the data detection time interval from the previous detection time t _m-1 to the current detection time t _m .

Further, the predicted detection time interval T ₂ from the current detection time t _m of the point A to the predicted detection time t _{m + 1} is T ₂ = t _{m + 1} −t _m = (2 · θ max −θ _{k -θ B · t m) /} (θ B + θ T) ... it is (5). On the other hand, the above equation _{(1) (t m, θ} k) Substituting, the _{θ k = θ B t m t} m = θ k / θ B. Therefore, in this case, the predicted detection time interval T ₂ can be expressed as T ₂ = 2 (θmax −θ _k ) / (θ _T + θ _B ) ... (5) ′.

Next, the case where the preceding vehicle is detected by the change in the reverse direction will be described. In FIG. 6, the reverse change locus indicated by # 1 ′, the forward change locus indicated by # 2 ′, and the # 3. change the trajectory of the angular position of the preceding vehicle shown _{is, θ = -θ B · t +} θmax ... (6) θ = θ B · t-θmax ... (7) θ = θ T · t + θ k + θ T · (-t m) ... It can be expressed as (8). Then, according to the above equations (7) and (8), the prediction detection time (t = t
_{m + 1)} is _{solved, t m + 1 = (θ} T · t m -θmax -θ k) / (θ T -θ B) ... it is (9).

Further, the prediction detection time interval T ₂ is given by T ₂ = t _{m + 1} −t _m = (− θ max −θ _k + t _m · θ _B ) / (θ _T −θ _B ) ... (10) . Here, the above equation _{(6) (t m, θ} k) _{Substituting, θ k = -θ B · t} m + θmax t m = (θmax -θ k) / θ B Therefore, in this case, The predicted detection time interval T ₂ can be expressed as T ₂ = 2 · θ _k / (θ _T −θ _B ) ... (10) ′.

Further, the future predicted distance r.pt and the future predicted angle θ. As predicted detected positions in the predicted position section 3.
Explaining the calculation processing of pt, they can be expressed by the following equations (11) and (12) using the data detection time interval T ₁ and the predicted detection time interval T ₂ . r.pt = (relative speed of the preceding vehicle) × T ₂ + (current distance from the preceding vehicle) (11) θ.pt = (angular speed of the preceding vehicle) × T ₂ + (current angle of the preceding vehicle) (( 12)

At this time, a smoothing distance r.st _obtained by smoothing the detection distance r _k and the detection angle θ _k by filtering the detection signal output from the laser head unit 1 to remove noise and disturbance. And the smooth angle θ.
α, which is applied to a radio wave radar, using st and a smooth distance differential value r.dt and a smooth angle differential value θ.dt (distance and angle change speed) obtained by differentiating them
By using the tracker, it is possible to make more accurate predictions.

The relationship among the predicted distance r.pt, the detected distance r _k , the smoothed distance r.st, and the smoothed distance differential value r.dt is as follows. r.pt (current predicted distance) = r.st (previous smoothed distance) + r.dt (previous smoothed distance differential value) × T ₁ (13) r.st (smoothed distance) = r.pt (current predicted distance) + Α {r _k (detection distance) -r.pt (current predicted distance)} (14) r.dt (smooth distance differential value) = r.dt (previous smooth distance differential value) + β {r _k (detection distance) −r.pt (current predicted distance)} / T ₁ (15) Therefore, the above equation (11) is r.pt (future predicted distance) = r.st (current smoothed distance) + r.dt (smoothed distance) (Differential value) × T ₂ (11) ′.

On the other hand, the relationship among the predicted angle θ.pt, the detected angle θ _k , the smooth angle θ.st, and the smooth angle differential value θ.dt is as follows. θ.pt (current predicted angle) = θ.st (previous smoothed angle) + θ.dt (previous smoothed angle differential value) × T ₁ (16) θ.st (smoothed angle) = θ.pt (current predicted angle) + Α {θ _k (detected angle) -θ.pt (current predicted angle)} (17) θ.dt (smooth angle differential value) = θ.dt (previous smooth angle differential value) + β {θ _k (detected angle) −θ.pt (current predicted angle)} / T ₁ (18) Therefore, the above formula (12) is θ.pt (future predicted angle) = θ.st (current smoothed angle) + θ.dt (smoothed angle) (Differential value) × T ₂ (12) ′.

Further, in the present embodiment, the control unit C causes the laser head unit 1 to detect the preceding vehicle in a predetermined area next time based on the predicted detection time calculated by the time predicting section 2. A detection determination unit 5 as a detection determination unit that determines whether or not it is possible,
When the detection determination unit 5 determines next time that the laser head unit 1 cannot detect the preceding vehicle within the predetermined region, the preceding vehicle is excluded from the comparison targets of the same determination performed by the same determination unit 4. And a management canceling unit 6 as a management canceling unit.

Specifically, if the positional relationship between the area which is the coverage area of the radar beam and the preceding vehicle is relatively narrow with respect to the angular velocity and the relative speed, if it deviates outside the area and cannot be detected next time. When it is predicted, the management of the preceding vehicle is stopped to reduce the management target number, whereby the processing load can be reduced. For example, in the example shown in FIG. 7, it is predicted that the preceding vehicle cannot be detected at the fifth change timing. Therefore, it is out of the management target at the end of the fourth change in the outgoing direction.

Here, the obstacle detection processing according to the present embodiment performed in the controller unit C will be described.
Description will be given based on the flowcharts of FIGS. First, in step S1, a maximum of n by the laser unit 1
A set of detection data (t _N1 , r _N1 , θ _N1 ), (t _N2 , r _N2 ,
θ _N2 ), ... (t _Nk , r _Nk , θ _Nk ) ... (t _Nn , r _Nn ,
θ _Nn ) is taken in and N table data is created. However, t _Nk
Is the detection time, r _Nk is the distance, and θ _Nk is the angle. In step S2, the M table data or the N table data, which is the current detection data for checking the correlation, is updated according to the situation, and then the process proceeds to step S3. It should be noted that the M table is the actual detection time t _Mk and the predicted detection time t.p _Mk for each target expected number,
Predicted distance r.pt, smoothed distance r.st and smoothed distance derivative r.
It is a collection of data consisting of dt, a predicted angle θ.pt, a smooth angle θ.st, and a smooth angle differential θ.dt, and is initialized in steps S23 to S25 described later immediately after the start of this processing. Become.

In step S3, it is determined whether the predicted detection time t.p _Mk of one preceding vehicle in the M table data is close to the actual detection time t _Nk of any preceding vehicle in the N table data. . That is, the predicted detection time t.p _Mk and the actual detection time t _Nk
Performs identification based on comparison with. When the determination is YES, the process proceeds to the next step S4, while when the determination is NO, the process returns to step S2. Then, in step S4, the actual detection time t _Nk closer to the predicted detection time t.pMk is selected, and then the process proceeds to step S5.

In step S5, the predicted distance r.pt is close to the current distance r _k and the predicted angle θ.pt is the current angle θ for the preceding vehicle selected in step S4.
Determine if it is close to _k . That is, the prediction distance r.pt and the prediction angle θ.pt are compared with the current distance r _k and the current angle θ _k to perform identification. If the determination is YES, the process proceeds to the next step S6, while if the determination is NO, the process returns to step S2. Then, the above step S6
Then, after selecting (r _k , θ _k ) closer to the predicted distance r.pt and the predicted angle θ.pt, the process proceeds to step S7.
Here, the steps S3 and S5 constitute the same determination unit 4.

In step S7, (r of the N table data is
_It is determined whether the investigation is completed up to _n , θ _n ). When the determination is YES, the process proceeds to the next step S8, while N
When it is O, the process returns to the step S2 and the above process is performed M
Repeat for every detection data in the table data. Then, in the above step S8, it is determined whether or not there is N table data having a correlation, and when the determination is YES, the step S9 in FIG. 9 is performed, and when NO, the step S in FIG.
Move to 20 respectively.

In step S9, the predicted angle θ.pt of the M table data, the current angle θ _k of the N table data selected to approximate the predicted angle θ.pt, and the N table data are detected. The data detection time interval T ₁ is calculated from the beam scanning direction at that time, and then the process proceeds to step S10.

In step S10, the current predicted distance r.pt, smoothed distance r.st and smoothed distance differential value r of the M table data are detected using the detected distance r _k of the N table data and the data detection time interval T _1. .dt is respectively calculated by the following equations (13) to (15) and rewritten, and the process proceeds to step S11. r.pt = r.s.t + r.dt × T ₁ (13) r.st = r.p.t + α (r _k −r.pt) (14) r.dt = r.d.t + β (r _{k -r.pt) / T 1 ... (} 15)

In step S11, the current prediction angle θ.pt, smoothing angle θ.st and smoothing angle differential value θ of the M table data are calculated using the detected angle θ _k of the N table data and the data detection time interval T _1. .dt is calculated by the following equations (16) to (18) and rewritten, and then the process proceeds to step S12. θ.pt = θ.s.t + θ.dt × T ₁ (16) θ.st = θ.p.t + α (θ _k −θ.pt)… (17) θ.dt = θ.d.t + β (θ _k− θ.pt) / T ₁ … (18)

In step S12, it is determined whether or not the preceding vehicle has been detected during forward scanning when creating the N table data. When the determination is YES, the process proceeds to steps S13 and S14, while when the determination is NO,
That is, if the laser beam is detected while being scanned in the reverse direction, the process proceeds to steps S15 and S16.

In step S13, the future prediction detection time t.p _Nk as the prediction detection time is obtained according to the following equation (4) described above, and t.p _Nk = (θ _T · t _Nk +2 · θ max _{-θ Nk) / (θ B +} θ T) ... (4) in the next step S14, obtains a predicted detection time interval T ₂ according to the following equation (5) ', then, step S1 of FIG. 10
Move to 7. _{T 2 = 2 (θmax -θ k} ) / (θ T + θ B) ... (5) '

On the other hand, in step S15, the future prediction detection time t.p _Nk is calculated by the following equation (9). That is. The time prediction unit 2 is configured by the steps S13 and S15.

[0053] _{t.p Nk = (θ T · t} Nk -θmax -θ Nk) / (θ T -θ B) ... (9) Then, in the next step S16, the prediction detected by the following equation (10) 'Time After obtaining the interval T ₂ , the process proceeds to step S17.

T ₂ = 2 · θ _Nk / (θ _T −θ _B ) ... (10) ′

In step S17, it is determined whether the predicted detection time interval T ₂ exceeds 0 and is less than one scanning time T _S of the laser beam (0 <T ₂ <T _S ).
This step S17 constitutes the detection determination unit 5,
It is determined whether or not the preceding vehicle is present within the detectable range at the next time. If the determination is YES, the process proceeds to steps S18 and S19, while if the determination is NO, the process proceeds to step S20.

First, at step S18, the future predicted distance r.pt is calculated by the above equation (11) 'using the current smoothed distance r.st, the smoothed distance differential value r.dt and the predicted detection time interval T _2. Is calculated. Next, in step S19, the future predicted angle θ.pt is calculated from the above equation (12) ′ using the current smoothed angle θ.st, the smoothed angle differential value θ.dt, and the predicted detection time interval T ₂ . Then, it moves to step S21. That is, the position prediction unit 3 is configured by these steps S18 and S19.

On the other hand, in step S20, the registration of the corresponding M table data is deleted. That is, the above step S
In the case of shifting from 17, only the registration of the data of the preceding vehicle whose predicted detection time interval T ₂ does not satisfy the condition of 0 <T ₂ <T _S is deleted. This step S20 constitutes the management canceling unit 6. Further, when the process moves from step S8, all the registrations of the M table data are deleted, and then the process moves to step S21.

In step S21, it is determined whether or not all the correlation investigations on the M table data have been completed. The judgment is YE
If S, the process proceeds to step S22 of FIG. 11, while N
When it is O, the process returns to step S2. Then, in step S22, it is determined whether or not a preceding vehicle is newly detected. If the determination is YES, step S23-
On the other hand, if NO in S25, the entire process is terminated.

In steps S23 to S25, the M table data is initialized. That is, in step S23, the detection time t _Nk of all the new preceding vehicles is fetched as data. In step S24, the detection distance r _Nk
Based on the above, each value of the initial predicted distance r.pt, the initial smoothed distance r.st, and the initial smoothed distance derivative r.dt is set.
Specifically, the initial predicted distance r.pt and the initial smoothed distance r.pt.
Since there is no detection data up to that point in st, the detection distance r _Nk is directly substituted as temporary data. Further, 0 is substituted into the initial smoothed distance differential r.dt for the same reason. Next, in step S25, each value of the initial predicted angle θ.pt, the initial smoothed angle θ.st, and the initial smoothed angle differential θ.dt is set based on the detected angle θ _Nk . After that, the process returns to step S22, and if there is no newly detected preceding vehicle, all the processes are finished.

Therefore, according to this embodiment, next time,
The predicted detection time t.p _Nk (→ t.p _Mk ) of the preceding vehicle that will be detected by the radar head unit 1 is calculated by the time prediction unit 2 and the next time the preceding vehicle is detected. Since the two preceding vehicles are identified based on the comparison with the actual detection time t _Nk , the comparison target of the identification is the preceding vehicle detected at the predicted detection time t.p _Mk. It is possible to reduce the processing load required for identification by narrowing down to, and thus it is possible to process obstacle data quickly and accurately.

In addition, the predicted distance r. Of the preceding vehicle which will be detected by the radar head unit 1 next time.
pt and the angle θ.pt are calculated by the position prediction unit 3, and when the preceding vehicle is detected next time, the identification of both preceding vehicles is made based on the comparison with the actual distance r _Nk and the angle θ _Nk . Since the comparison target of the identification is further narrowed down to the preceding vehicle existing at the predicted detection position, the processing load required for the identification can be further reduced.

The predicted distance r.pt and the angle θ.pt
Since the predicted detection time interval T ₂ derived from the predicted detection time t.p _Nk is used in the calculation of, the detection time t for each preceding vehicle even at the same transmission direction change timing. Despite the variations in _Nk , the predicted distance r.pt and the angle θ.pt with high accuracy can be obtained for each preceding vehicle.

Further, based on the predicted detection time interval T ₂ , the detection determination unit 5 determines whether or not the next preceding vehicle can be detected within the predetermined area, and the detection is impossible. Therefore, the management of the preceding vehicle is stopped, so that the processing load required for the same determination for the preceding vehicle determined to be undetectable within the predetermined area at the time of the next detection by the radar head unit 1 is eliminated. Therefore, the processing load of the obstacle detection device can be reduced.

In the above embodiment, the same determination is made based on the comparison between the predicted detection time and the actual detection time as well as the comparison between the predicted detection position and the actual detection position.
It may be performed based on only the comparison between the predicted detection time and the actual detection time or only the comparison between the predicted detection position and the actual detection position.

Further, in the above-described embodiment, the detection device for detecting the object by reciprocally changing the laser beam in the horizontal direction has been described, but the oscillating wave is not limited to the laser beam, and the oscillation direction can be changed. The form may also have a certain change characteristic.

[0066]

As described above, according to the first aspect of the invention, the transmitting section for transmitting the detected wave, the receiving section for receiving the reflected wave from the object of the detected wave, and the transmitting direction of the detected wave. Is composed of a transmission direction changing unit for changing the transmission direction so as to change within a predetermined range at a predetermined angular velocity, and is provided with an object detection means for detecting an object existing in a predetermined area, and is detected by the object detection means. In an obstacle detection device configured to detect an obstacle based on the displacement amount and the direction of displacement of the object within a predetermined time, this time, the displacement characteristic of the object detected by the object detection means and the transmission of the object detection means Based on the change characteristic of the direction, the time prediction means for calculating the predicted detection time of the object detected by the object detection means next time, the predicted detection time calculated by the time prediction means, and the object for the next time. Based on the comparison with the actual detection time when is detected, the same determination means for determining whether the object detected this time and the object detected the next time are the same Therefore, the comparison target of the identification can be narrowed down to the objects detected at the predicted detection time, and the processing load of the CPU required for the identification can be reduced accordingly, and the obstacle data can be obtained. You will be able to process quickly and accurately.

According to the invention of claim 2, based on the displacement characteristic of the object detected this time by the object detecting means,
In addition to the position prediction means for calculating the predicted detection position of the object detected by the object detection means at the next time, the same determination means is added to the comparison between the predicted detection time and the actual detection time. Since the identification is performed based on both the comparison between the predicted detection position calculated by the position prediction means and the actual detection position when the object is detected next time, the comparison target of the identification is predicted. The objects existing at the detection position can be further narrowed down, so that the processing load required for the identification can be further reduced.

According to the invention of claim 3, the position predicting means is configured to calculate the predicted detection position of the object detected at the predicted detection time calculated by the time predicting means, and the prediction thereof is performed. Since it is made the same based on the comparison between the detected position and the actual detected position, it is possible to accurately predict the detected position of the object detected at the next time and to correct the obstacle data processing. You can

According to the fourth aspect of the present invention, the same determining means compares the predicted detection position calculated based on the predicted detection time with the actual detection position, and further compares the predicted detection time with the actual detection time. Since the identification is performed based on both comparisons, the processing load for the identification can be further reduced.

According to the fifth aspect of the present invention, the detection determination means for determining whether or not the object can be detected within the predetermined area at the next time based on the predicted detection time, and the detection determination means. Since it is equipped with a management stop means for stopping the management of the object when it is determined that the next detection of the object in the predetermined area is impossible by the means, the next time the object cannot be detected in the predetermined area. It is possible to eliminate the processing load required for the same determination for the object determined to be
It is possible to contribute to the reduction of the processing load of the obstacle detection device.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an obstacle detection device according to the present invention.

FIG. 2 is a perspective view showing an automobile equipped with an obstacle detection device according to an embodiment of the present invention.

FIG. 3 is a block diagram showing an overall configuration of an obstacle detection device including peripheral devices.

FIG. 4 is an angle-time characteristic diagram for explaining a predicted detection time interval T ₂ of the preceding vehicle detected when the laser beam in the forward direction changes.

FIG. 5 is an angle-time characteristic diagram for explaining the angular velocity of the preceding vehicle and the data detection time interval T ₁ obtained based on the angle and time at the time of previous detection and the angle and time at the time of current detection.

FIG. 6 is a diagram corresponding to FIG. 4 for explaining a predicted detection time interval T ₂ of the preceding vehicle detected when the laser beam in the opposite direction changes.

FIG. 7 is an angle-time characteristic part showing the state of the preceding vehicle deviating from the detection area.

FIG. 8 is a flowchart showing an initial stage part of obstacle detection processing.

FIG. 9 is a flowchart showing a second stage portion of obstacle detection processing.

FIG. 10 is a flowchart showing a third stage portion of obstacle detection processing.

FIG. 11 is a flowchart showing the final stage of the obstacle detection process.

FIG. 12 is a schematic diagram showing displacements of a plurality of preceding vehicles in the scanning obstacle detection device for each beam change.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Radar head unit (object detection means) 1a Transmission section 1b Reception section 1c Transmission direction change section 2 Time prediction section (time prediction section) 3 Position prediction section (position prediction section) 4 Same determination section (same determination section) 5 Detection determination Department (detection determination means) 6 Management discontinuation department (management cancellation means)

Claims (5)

[Claims] 1. A transmitting unit for transmitting a detected wave, a receiving unit for receiving a reflected wave of the detected wave from an object, and a transmitting direction for changing the transmitting direction of the detected wave within a predetermined range at a predetermined angular velocity. And an object detection means for detecting an object existing in a predetermined area, and based on the displacement amount and the displacement direction of the object detected by the object detection means within a predetermined time. In the obstacle detection device configured to detect an obstacle, the object detection means detects the next time based on the displacement characteristics of the object detected by the object detection means and the change characteristics of the transmission direction. A time prediction unit that calculates a predicted detection time of an object, a predicted detection time calculated by the time prediction unit, and an actual detection time when the object is detected by the object detection unit next time. On the basis of the comparison, the obstacle detection is characterized in that the object detected this time by the object detection means and the same judgment means for judging whether the object detected next time are the same or not. apparatus. 2. The obstacle detection device according to claim 1, wherein the predicted detection position of the object detected next time by the object detection means is calculated next time based on the displacement characteristic of the object detected by the object detection means this time. The same determination means is based on the comparison between the predicted detection time and the actual detection time, and in addition to the predicted detection position calculated by the position prediction means, the next time the object detection means detects the object. Is configured to determine whether or not the object detected this time by the object detection means and the object detected next time are the same, based on the comparison with the actual detection position when is detected. An obstacle detection device characterized by being present. 3. A transmitting unit for transmitting a detected wave, a receiving unit for receiving a reflected wave of the detected wave from an object, and a transmitting direction for changing the transmitting direction of the detected wave within a predetermined range at a predetermined angular velocity. And an object detection means for detecting an object existing in a predetermined area, and based on the displacement amount and the displacement direction of the object detected by the object detection means within a predetermined time. In the obstacle detection device configured to detect an obstacle, the object detection means detects the next time based on the displacement characteristics of the object detected by the object detection means and the change characteristics of the transmission direction. Time prediction means for calculating the predicted detection time of the object, and this time, based on the displacement characteristics of the object detected by the object detection means, the detection is performed at the predicted detection time calculated by the time prediction means. And a predicted position detected by the position predicting unit, and a real detected position when the object is detected next time by the object detecting unit. Based on the above, the obstacle detection device characterized by further comprising: an identical determination means for determining whether or not the object detected this time by the object detection means and the object detected next time are the same. . 4. The obstacle detection device according to claim 3, wherein the same determination means is based on a comparison between the predicted detection position and the actual detection position, and the predicted detection time calculated by the time prediction means and the next time. , It is determined whether the object detected this time by the object detection means is the same as the object detected next time based on the comparison with the actual detection time when the object is detected by the object detection means. An obstacle detection device, which is configured to: 5. The obstacle detection device according to claim 1, 2, 3 or 4, wherein based on the predicted detection time calculated by the time prediction means,
Next time, the detection determination means for determining whether or not the object detection means can detect the object within the predetermined area, and the detection determination means for the next time cannot detect the object within the predetermined area by the object detection means. When it is determined that there is,
An obstacle detection device comprising: a management stopping unit that excludes the object from a comparison target of the same determination made by the same determination unit.