JP7230894B2 - self-driving device - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic driving device that transitions a vehicle to one of a plurality of driving modes.

It has been proposed to assess the safety of routes traveled by vehicles. For example, in Patent Document 1, a plurality of route candidates for a vehicle are generated, the behavior of the vehicle is predicted for the plurality of route candidates generated, and one of the route candidates is selected by evaluating the prediction results. is described (for example, Patent Document 1).

JP 2019-137321 A

The technique described in Patent Literature 1 has a problem that it takes an excessive amount of time to select a route candidate because it is necessary to evaluate each of the plurality of route candidates generated.

Therefore, the present invention has been made in view of these points, and an object of the present invention is to provide a technique capable of shortening the time required to select an appropriate travel route for a vehicle.

An automatic driving device according to a first aspect of the present invention includes an acquisition unit that acquires state information indicating a state of a vehicle and environment information indicating a state of a surrounding environment of the vehicle; a mode specifying unit that specifies the current running mode of the vehicle among the running modes of the vehicle; a candidate selection unit that selects candidates for the running mode of the vehicle; a prediction unit that predicts a future state of the surrounding environment based on the environment information; the state information; and the future surroundings predicted by the prediction unit. determining whether or not the vehicle satisfies a transition condition for transitioning the vehicle from the driving mode identified by the mode identifying unit to the driving mode selected by the candidate selecting unit based on the state of the environment and the state of the environment; and a transition unit that transitions the vehicle to the driving mode selected by the candidate selection unit when the determination unit determines that the vehicle satisfies the transition condition.

The determination unit estimates a collision risk when the vehicle travels in the driving mode selected by the candidate selection unit based on the state of the future surrounding environment predicted by the prediction unit, and determines the collision risk. Based on this, it may be determined whether the vehicle satisfies the transition condition. The determination unit estimates the length of a travel route traveled by the vehicle in the travel mode selected by the candidate selection unit, and determines whether the vehicle satisfies the transition condition based on the estimated length of the travel route. It may be determined whether

The acquisition unit acquires the environment information including information indicating traffic conditions around the vehicle, and the candidate selection unit selects a plurality of driving modes other than the driving mode specified by the mode specifying unit based on the traffic conditions. A priority may be assigned to each of the running modes, and some of the running modes may be selected as candidates based on the assigned priority. The transition unit may transition the vehicle to the driving mode selected by the candidate selection unit from among the plurality of driving modes corresponding to the combinations of the plurality of lanes and the plurality of target speeds. The automatic driving device further includes an operation identification unit that identifies an operation of the driver of the vehicle, and the candidate selection unit selects a part of the driving modes based on the operation identified by the operation identification unit. Can be selected as a candidate. The candidate selection unit may select a new candidate for the driving mode at predetermined time intervals.

ADVANTAGE OF THE INVENTION According to this invention, it is effective in shortening the time required for selecting the suitable driving|running route of a vehicle.

BRIEF DESCRIPTION OF THE DRAWINGS It is a schematic diagram for demonstrating an example of a structure of the automatic driving apparatus which concerns on 1st Embodiment. It is a figure which shows the structure of a control apparatus. FIG. 4 is a schematic diagram for explaining determination of a driving behavior in short-term prediction by an identification unit and a travel route identified in long-term prediction; FIG. 5 is a schematic diagram for explaining an example of a travel route specified by an specifying unit; It is a figure which shows the example of several driving modes. FIG. 10 is a diagram showing another example of vehicle running modes; It is a figure for demonstrating the calculation method of the function value of a cost function by a calculation part. It is a figure which shows the example of determination of the effectiveness of a lane change by a selection part. FIG. 5 is a diagram showing an example of lane change end determination by a steering control unit; It is a flow chart which shows a processing procedure of selection of a lane which vehicles run by automatic driving device 1 of a 1st embodiment. It is a figure which shows the structure of the control apparatus with which the automatic driving apparatus of 2nd Embodiment is provided. It is a figure which shows the example of several driving modes. FIG. 5 is a diagram showing an example of selection of driving mode candidates by a candidate selection unit; It is a flowchart which shows the processing procedure of the transition of driving modes by the automatic operation apparatus of 2nd Embodiment.

<First embodiment>
[Configuration of automatic driving device]
A configuration of an automatic driving device according to a first embodiment of the present invention will be described with reference to FIG.

FIG. 1 is a schematic diagram for explaining an example of the configuration of an automatic driving device 1 according to the first embodiment. The automatic driving device 1 is mounted on a vehicle such as a truck, for example, and assists the driving of the vehicle. The automatic driving device 1 includes a vehicle detection unit 2, an environment recognition unit 4, a map database 6, and a control device 10, as shown in FIG.

The vehicle detection unit 2 detects the state of the vehicle. The vehicle detection unit 2 detects the position and speed of the vehicle and whether or not the direction indicator is blinking. For example, the vehicle detection unit 2 has a GPS (Global Positioning System) receiver, and detects the position of the vehicle from radio waves received by the GPS receiver. The vehicle detection unit 2 outputs state information indicating the detection result of the state of the vehicle to the control device 10 .

The environment recognition unit 4 recognizes the state of the surrounding environment of the vehicle. For example, the environment recognition unit 4 has an external sensor such as a camera or radar. The environment recognition unit 4 recognizes the position, moving direction, moving speed, and acceleration of objects around the vehicle (for example, other vehicles, bicycles, pedestrians, etc.) based on the output of the external sensor. In addition, the environment recognition unit 4 can recognize traffic conditions such as the position and width of the lane in which the vehicle is traveling, whether there are other lanes on the left and right of the vehicle, and whether another vehicle is traveling in front of the vehicle. . The environment recognition unit 4 outputs the result of recognizing the state of the surrounding environment of the vehicle and the result of recognizing the traffic situation to the control device 10 as environment information.

The map database 6 stores road map information. The road map information includes, for example, data indicating three-dimensional coordinates of latitude, longitude and altitude of roads. The road map information also includes information on the number of lanes and lane structure of the road on which the vehicle travels. In addition, the map database 6 can also acquire the traffic conditions recognized by the environment recognition unit 4 based on the position of the vehicle detected by the vehicle detection unit 2 instead.

The control device 10 controls the operation of the automatic driving device 1 . Control device 10 identifies a travel route on the assumption that the vehicle travels in each of a plurality of lanes including the lane on which the vehicle is currently traveling. The control device 10 estimates the collision risk on each of the specified travel routes. Although the details will be described later, the control device 10 calculates the function value of the cost function for evaluating the travel route for each travel route based on the estimated collision risk of the travel route and the length of the travel route. calculate. The control device 10 selects the travel route that minimizes the calculated function value as the travel route on which the vehicle travels. In this manner, the control device 10 can automatically determine whether a lane change is necessary or not and automatically select a lane to change to without the driver instructing whether or not to change lanes.

However, if the travel route with the smallest function value is selected from among the plurality of travel routes, the frequency of lane changes by the vehicle may become excessive depending on the traffic conditions around the vehicle. Therefore, the control device 10 determines the necessity of changing lanes using a cost function including adjustment values corresponding to each of the plurality of travel routes, thereby adjusting the frequency with which each travel route is selected. In this way, control device 10 can prevent the frequency of lane changes from becoming excessive.

[Configuration of control device]
FIG. 2 is a diagram showing the configuration of the control device 10. As shown in FIG. The control device 10 includes a storage section 11 and a control section 12 . The control unit 12 includes an identification unit 121 , a calculation unit 122 , a selection unit 123 and a steering control unit 124 .

The storage unit 11 includes a storage medium such as a ROM (Read Only Memory), a RAM (Random Access Memory), and a hard disk. The storage unit 11 stores programs executed by the control unit 12 . The control unit 12 is, for example, a CPU (Central Processing Unit). The control unit 12 functions as an identification unit 121 , a calculation unit 122 , a selection unit 123 and a steering control unit 124 by executing programs stored in the storage unit 11 .

[Identification of driving route]
The identifying unit 121 identifies a travel route along which the vehicle travels. The identification unit 121 determines the driving behavior of the vehicle using, for example, the potential field, and generates a travel route for the vehicle to travel in the lane based on the determined driving behavior. A potential field is calculated|required by a well-known method, for example. The identifying unit 121 determines a driving behavior based on short-term prediction and generates a driving route based on long-term prediction. The travel route is the trajectory of the vehicle position during long-term prediction.

An outline of route generation in long-term prediction in this embodiment will be described with reference to FIG.
FIG. 3 is a schematic diagram for explaining the determination of the driving behavior in the short-term prediction by the identification unit 121 and the travel route identified in the long-term prediction. As shown in FIG. 3, the short-term forecast is the forecast performed during the period _Δts , and the long-term forecast is the forecast performed during the period _ΔtL . As shown in FIG. 3, the identifying unit 121 performs short-term prediction for the period _Δts a plurality of times (here, four times) during the period _ΔtL . At this time, the driving behavior determined by one short-term prediction is reflected in the next short-term prediction. For example, the identifying unit 121 performs the short-term prediction P2 in consideration of the vehicle state and the positions of surrounding objects that reflect the driving behavior determined by the short-term prediction P1. The identification unit 121 identifies the travel route by connecting the trajectory of the vehicle position based on the driving behavior determined by the short-term prediction described above. At this time, the identification unit 121 identifies the travel route to which the collision risk information is added based on the potential method, the Kalman filter, or the like. Note that the number of short-term prediction execution times does not necessarily have to be the number obtained by dividing the period Δt _L by the period Δt _s . That is, the prediction time Δt _L of short-term prediction may be longer or shorter than the time difference between each route point.

FIG. 4 is a schematic diagram for explaining an example of a travel route specified by the specifying unit 121. As shown in FIG. The travel route shown in FIG. 4 includes a plurality of route points R1 to R4. Path points R1-R4 are the positions of the vehicle estimated by short-term prediction. Point R0 is the starting point for long-term prediction. The route point R1 is the position of the own vehicle determined by the optimum driving behavior (that is, acceleration/deceleration and yaw rate) determined by the short-term prediction at the point R0. Route point R2 is the position of the vehicle determined by the optimal driving behavior determined by the short-term prediction at route point R1. Similarly, path point R3 is the position of the ego vehicle determined by the optimal driving behavior determined by the short-term prediction at path point R2. Here, in each short-term prediction that is repeatedly performed, driving behavior is determined so as to ensure a safe inter-vehicle distance with respect to the predicted positions of surrounding objects.

The identification unit 121 can generate a route that takes into account changes in traffic conditions at each route point by identifying a travel route that connects the route points R1 to R4, thus making the route safer. can be specified. Note that the dashed arrows in FIG. 4 indicate driving behaviors that are not optimal in short-term prediction.

The identifying unit 121 estimates a collision risk on each of the identified travel routes. More specifically, the identifying unit 121 predicts the position and speed of the surrounding obstacle at a plurality of future prediction times based on the speed of the surrounding obstacle, or the speed and acceleration of the surrounding obstacle. do. The identifying unit 121 calculates the collision risk between the surrounding obstacle and the vehicle based on the predicted position and speed of the surrounding obstacle. The identification unit 121 may terminate the generation of the travel route in the middle. For example, when the collision risk of a route point on the travel route is equal to or greater than a predetermined threshold value R _th ^coll , the specifying unit 121 specifies the travel route from the start point to the route point immediately before this route point, and The travel route after the route point is not specified. As a result, it is possible to suppress generation of a route whose safety is not sufficiently ensured.

Note that the identification unit 121 may identify the travel route without stopping the identification of the travel route even when the collision risk is equal to or greater than the predetermined threshold value R _th ^coll , without being limited to the above. In this case, it is desirable to notify the driver that there is a collision risk on the identified travel route. Thereby, the identification unit 121 can make the driver recognize the collision risk when the host vehicle automatically drives along the identified travel route.

The specifying unit 121 specifies a plurality of travel routes when the vehicle travels in each of a plurality of lanes including the lane on which the vehicle is currently traveling. FIG. 5 is a diagram showing an example of multiple running modes. The example of FIG. 5 shows three driving modes, a standard left lane change mode, a standard lane keeping mode, and a standard right lane change mode. The standard left lane change mode is a driving mode for causing the vehicle to change lanes at the same target speed as the current lane to the lane on the left of the current lane. The standard lane keeping mode is a driving mode for driving the vehicle in the current lane without changing lanes at the same target speed as the current one. The standard right lane change mode is a driving mode in which the automatic driving device 1 causes the vehicle to change lanes to the lane on the right of the current lane at the same target speed as the current one.

The identifying unit 121 identifies driving routes corresponding to the driving modes of the standard left lane change mode, the standard lane maintenance mode, and the standard right lane change mode. The identifying unit 121 identifies, as a driving route corresponding to the standard left lane change mode, a driving route for changing lanes to a lane adjacent to the left lane of the current driving lane. The identification unit 121 identifies, as a travel route corresponding to the standard travel maintenance mode, a travel route in which the vehicle travels in the current lane without changing lanes. The identification unit 121 identifies, as the standard right lane change mode, a traveling mode in which the vehicle is currently traveling in a lane that is adjacent to the right on the right side of the lane.

Further, the identification unit 121 is not limited to the example of identifying the travel routes corresponding to the three travel modes. For example, the identification unit 121 may identify travel routes corresponding to a plurality of travel modes in which the vehicle travels at different target speeds. In addition, the identifying unit 121 may identify driving routes corresponding to a plurality of driving modes that combine a plurality of target speeds and a plurality of lanes on which the vehicle travels. FIG. 6 is a diagram showing another example of multiple running modes.

In FIG. 6, the standard left lane change mode, the standard lane maintenance mode, and the standard right lane change mode are the same as the examples shown in FIG. The identifying unit 121 identifies, as a traveling route corresponding to the high speed left lane change mode, a traveling route in which the vehicle is to change lanes in the lane adjacent to the left of the current lane at a higher target vehicle speed than in the standard left lane change mode. The identifying unit 121 identifies, as a driving route corresponding to the low-speed left lane change mode, a driving route in which the vehicle is to change lanes in the lane adjacent to the left of the current lane at a target vehicle speed lower than that in the standard left lane change mode. The same applies to the fast right lane change mode and the slow right lane change mode. The target speed for each driving mode shall be specified in advance by the driver or the automatic driving system.

[Conditions of driving route]
The specifying unit 121 specifies a travel route in which collision risks R _opt,j ^coll at a plurality of route points on the travel route are all smaller than a predetermined threshold value Rthcoll (R _opt,j ^coll <R _th ^coll (equation (1)). ). The threshold value R _th ^coll is, for example, a value lower than the collision risk during driving of the vehicle by an average driver. The collision risk R _opt,j ^coll is obtained based on a known technique such as the potential method. j is the index number of the path point.

When the collision risk R _opt,j ^coll at any route point on the travel route is greater than or equal to the threshold value R _th ^coll , the specifying unit 121 determines whether the travel from the start point of the travel route to the route point immediately before this route point is determined. The route is specified, and the travel route beyond the route point where the collision risk Ropt,jcoll becomes equal to or greater than the threshold value Rthcoll is not specified. The identification unit 121 associates the identified travel route, the information indicating the travel mode corresponding to the travel route, and the information indicating the collision risk at a plurality of route points on the travel route, and outputs the information to the calculation unit 122 . The specifying unit 121 sequentially re-specifies the travel route corresponding to the current travel mode, and outputs the re-specified travel route to the steering control unit 124 .

[Calculation of function value of cost function]
The calculation unit 122 calculates a travel route including, as parameters, the length of each of the plurality of travel routes, the collision risk in each of the plurality of travel routes identified by the identification unit 121, and the adjustment value corresponding to each of the plurality of travel routes. A function value of a cost function for evaluating is calculated for each travel route. FIG. 7 is a diagram for explaining a method of calculating the function value of the cost function by the calculator 122. As shown in FIG.

In the example of FIG. 7, the identification unit 121 corresponds to the driving route corresponding to the standard left lane change mode in which the lane is changed to the adjacent lane on the left, and the standard lane keeping mode in which the current lane is driven without changing the lane. and a driving route corresponding to the standard right lane change mode in which the lane is changed to the adjacent lane on the right. The travel route identified by the identification unit 121 is indicated by an arrow. In FIG. 7, a plurality of route points on the travel route are indicated by black circles. A plurality of route points indicate positions where the vehicle is expected to exist on the travel route at predetermined time intervals.

n _end in FIG. 7 indicates the number of route points of each travel route. The time t _end is the predicted time when the vehicle reaches the last route point on the travel route specified by the specifying unit 121 . The identification unit 121 terminates the identification of the travel route when the collision risk R _opt,j ^coll at any route point on the travel route is greater than or equal to the threshold value R _th ^coll . t _end is the estimated time that the vehicle will reach the route point immediately before the route point to which the collision risk equal to or greater than the threshold value R _th ^coll is given if the identification of the travel route is terminated. The number of route points n _end of different travel routes in FIG. 7 may not be the same value. The same is true for the expected time t _end at which the vehicle will reach the last route point.

First, the calculation unit 122 obtains statistics of collision risks at a plurality of route points on the travel route identified by the identification unit 121 . For example, the calculation unit 122 obtains the maximum value R _max ^coll of the collision risk at a plurality of route points of the travel route shown in FIG. 7 for each travel route.

The calculation unit 122 calculates the function value C ^path (I) of the cost function for evaluating the travel route based on the obtained statistic. I is an index number representing the running mode. A cost function is represented by the following formulas (2) and (3), for example.
C ₀ ^path =−w _{S tend} /(V _target Δt _L )+w _R (R _max ^coll /R _th ^coll ). . . (2)
C ^path (I)=C ₀ ^path +C ^mode (I). . . (3)
In equation (2), the first term is normalized by _dividing the distance traveled distance st end by the ideal distance calculated as the product of the target speed V _target and the expected time Δt _L. is the value In equations (2) and (3), this first term is used as a gain, ie, a negative cost, to calculate the function value C ^path (I). The second term of the formula (2) indicates a penalty, ie, a positive cost, which is the normalized value of the risk when traveling on the travel route. In equation (2), the first term st _end indicates the length of the route that the vehicle is expected to travel from time t ₀ to time t _end on the travel route. Time _t0 is the prediction start time.

Δt _L is the expected time at which the long-term forecast is made. Assuming that Δt _end =t _end −t ₀ , Δt _L =Δt _end if the identification of the travel route by the identification unit 121 is not interrupted. V _target is the target vehicle speed in the current driving mode of the vehicle in the example of this specification, but may be the target vehicle speed in the driving mode for calculating the function value C ^path (I). w _S and w _R are weighting factors. When the weighting factor ws is set large, the magnitude of the gain in the first term is more likely to be evaluated than the magnitude of the penalty in the second term. When the weighting factor _wR is set large, the magnitude of the penalty in the second term is more likely to be evaluated than the magnitude of the gain in the first term.

In Equation (3), C ^mode (I) is an adjustment value that indicates the difficulty of transitioning to the running mode. For example, when calculating the function value C ^path (I) corresponding to the current running mode, C ^mode (I)=0. On the other hand, when calculating the function value C ^path (I) corresponding to a running mode different from the current running mode, C ^mode (I) becomes a value greater than zero. Increasing the adjustment value C ^mode (I) for the traveling mode involving lane changes reduces the frequency of lane changes. On the other hand, when the adjustment value C ^mode (I) for the travel mode that accompanies lane changes is decreased, the frequency of lane changes increases.

[Adjustment of adjustment value by driver's operation]
When the driver operates the left turn signal lamp, it can be inferred that the driver intends to change lanes to the adjacent lane on the left. Therefore, when an operation reception unit (not shown) receives a driver's operation to blink the direction indicator of the vehicle, the calculation unit 122 calculates the adjustment value C ^mode (I) may be reduced. In this way, the calculation unit 122 can facilitate selection of the travel route intended by the driver.

Further, when the driver operates a key or the like for increasing the target vehicle speed, it can be estimated that the driver intends the vehicle to travel at a higher target vehicle speed. When the operation receiving unit receives a driver's operation to increase the target vehicle speed of the vehicle, the calculating ^unit 122 calculates a relatively The travel route adjustment value C ^mode (I) corresponding to a high target vehicle speed may be decreased.

For example, when the operation reception unit receives a driver's operation to increase the target vehicle speed of the vehicle, the calculation unit 122 compares the adjustment value C ^mode (I) corresponding to the low-speed left lane change mode shown in FIG. , the adjustment value C ^mode (I) corresponding to the fast left lane change mode may be decreased. The calculator 122 calculates the function value C ^path (I) of the cost function for each travel route. Calculation unit 122 outputs information indicating the travel route and the function value C ^path (I) of this travel route to selection unit 123 .

[Determination of effectiveness of lane change]
The selection unit 123 selects one of the plurality of travel routes specified by the specifying unit 121 . First, the selection unit 123 determines whether or not the lane change included in the travel route corresponding to the travel mode selected by the candidate selection unit 525 is valid.

Selecting unit 123 selects the absolute value of the difference between the azimuth angle ψ _j ^ego of the traveling direction of the vehicle after the lane change and the azimuth angle θ of the direction in which the lane boundary line extends, for the travel route corresponding to the travel mode involving the lane change. is equal to or less than a predetermined reference value Δθ _th (|ψ _j ^ego −θ|≤Δθ _th (Equation 4)), it is determined that the lane change included in the travel route is valid. Selection unit 123 determines that the absolute value of the difference between the azimuth angle ψ _j ^ego of the traveling direction of the vehicle after the lane change included in the travel route and the azimuth angle θ in the direction in which the lane boundary line extends is a predetermined reference value Δθ _th If it is greater than (|ψ _j ^ego -θ|>Δθ _th (equation 5)), it is determined that the lane change included in this travel route is not valid, and the lane corresponding to this travel route is not selected. The reference value Δθ _th is, for example, the maximum absolute value of the difference between the two azimuth angles assumed when the vehicle is running stably in the lane.

The selection unit 123 determines whether the absolute value of the difference between the coordinate value of the route point on the travel route after the lane change and the coordinate value of the center of the lane after the lane change on the coordinate axis in the direction orthogonal to the lane boundary line is equal to or less than a threshold. determine whether or not FIG. 8 is a diagram showing an example of lane change effectiveness determination by the selection unit 123 .

The example of FIG. 8 shows a travel route for changing lanes to a lane on the right side of the lane in which the vehicle is currently traveling. A plurality of route points on the travel route are indicated by black circles. In FIG. 8, the X-axis is the coordinate axis in the extending direction of the lane boundary line, and the Y-axis is the coordinate axis in the direction orthogonal to the lane boundary line.

The selection unit 123 obtains the Y-axis coordinate value Y _j ^path of the route point A on the travel route after the lane change. The identifying unit 121 obtains the coordinate value Y _j ^cent in the Y-axis direction of the lane center A′ on the left side of the current lane. Here, the selection unit 123 determines whether or not the following formula (6) holds on the travel route after the lane change.
|Y _j ^path −Y _j ^cent |<U _th . . . (6)
In equation (6), U _th is a threshold. The threshold U _th is, for example, the maximum absolute value of the difference between two coordinate values assumed when the vehicle is completely in the left adjacent lane. The selection unit 123 does not select a point that does not satisfy Expression (4) or Expression (6) as the end point of the route. If the absolute value of the difference between the position of the vehicle in the direction perpendicular to the lane boundary after the lane change and the center of the lane after the lane change is greater than a threshold value, the selection unit 123 selects (|Y _j ^path −Y _j ^cent |≧U _th (7)), it is determined that the lane change included in this travel route is not valid, and the lane corresponding to this travel route is not selected. On the other hand, when the travel route satisfies the formulas (4) and (6), the selection unit 123 determines that the lane change included in the travel route is valid.

The selection unit 123 selects a lane along which the vehicle travels, based on the function value C ^path (I) of the cost function calculated by the calculation unit 122 . The selecting unit 123 selects the traveling route that does not involve a lane change and the traveling route that involves a lane change that is determined to be valid, and the function value C ^path (I) of the cost function calculated by the calculating unit 122 becomes the minimum. Select a driving route.

When only one travel route is determined to be valid for a lane change, the selection unit 123 selects a travel route that does not involve a lane change and one travel route that is determined to be valid for a lane change. By comparison, the travel route with the smaller function value C ^path (I) of the cost function is selected. Selection unit 123 selects a travel mode corresponding to the selected travel route. Selection unit 123 outputs information indicating the selected running mode to steering control unit 124 .

[Turn indicator lighting control]
The selection unit 123 blinks a direction indicator (not shown) based on the selection result of the travel route. For example, the selection unit 123 determines whether or not the left direction indicator 9 is on when a travel mode involving a lane change to a lane adjacent to the left of the current lane is selected. When the selection unit 123 determines that the left direction indicator is in the OFF state, the selection unit 123 causes the left direction indicator to transition from the OFF state to the ON state, and blinks the left lighting indicator.

When the lane change to the adjacent lane on the left is completed or interrupted, the selection unit 123 causes the lighting indicator to transition from the ON state to the OFF state, and extinguishes the left lighting indicator. When a predetermined period of time has elapsed since the lighting indicator was switched to the ON state in the standard lane keeping mode, the selection unit 123 switches the lighting indicator from the ON state to the OFF state and turns off the left lighting indicator. Let

[Transition of driving mode]
The steering control unit 124 controls steering of the vehicle. As shown in FIG. 5, the steering control unit 124 causes the vehicle to travel by transitioning to one of the standard left lane change mode, the standard lane maintenance mode, and the standard right lane change mode. In the standard left lane change mode, the steering control unit 124 changes lanes to the lane on the left of the lane in which the vehicle is currently traveling. In the standard lane keeping mode, the steering control unit 124 runs in the current lane without changing lanes. In the standard right lane change mode, the steering control unit 124 changes lanes to the lane on the right of the current lane.

The steering control unit 124 starts lane change to the selected lane based on the lighting state of the direction indicator on the lane side selected by the selection unit 123 . For example, when the selection unit 123 selects the standard left lane change mode, the steering control unit 124 determines that the left direction indicator is in an ON state and that a predetermined time or more has elapsed since the direction indicator was turned on. Then, the vehicle shifts to the left lane change mode and starts lane change to the adjacent lane on the left. At this time, the steering control unit 124 does not travel along the travel route itself used by the selection unit 123 to select the travel mode, but the travel route corresponding to the selected travel mode is newly specified by the identification unit 121. The vehicle shall travel along the revised travel route.

When the selection unit 123 selects the standard left lane change mode, the steering control unit 124 does not transition to the standard left lane change mode unless the left direction indicator is turned on. When the selection unit 123 selects the standard left lane change mode, the steering control unit 124 does not transition to the standard left lane change mode unless a predetermined period of time or more has elapsed since the left direction indicator turned on. .

[Determination of end of lane change]
After transitioning to the standard left lane change mode or the standard right lane change mode, the steering control unit 124 transitions to the standard lane keeping mode when the lane change is completed or interrupted. FIG. 9 is a diagram showing an example of lane change end determination by the steering control unit 124 . For example, if the entire vehicle is in the vicinity of the center of the lane to be changed, the steering control unit 124 ends the standard left lane change mode or the standard right lane change mode and transitions to the standard lane maintenance mode (see FIG. 9). (1)). As an example, when the vehicle satisfies the above formulas (4) and (6), the steering control unit 124 determines that the entire vehicle is in the vicinity of the center of the lane to be changed, and changes the standard left lane. mode or normal right lane change mode.

Further, when the environment recognition unit 4 in FIG. 1 identifies only the lane boundary line of the lane in which the vehicle is currently traveling, the steering control unit 124 detects when the front center of the vehicle is within the lane to be changed, that is, When the front of the vehicle crosses the lane boundary line, it may be determined that the lane change has ended ((2) in FIG. 9), and the standard left lane change mode or the standard right lane change mode may be terminated.

After the selection unit 123 selects the standard left lane change mode or the standard right lane change mode, the specifying unit 121 re-specifies the travel route corresponding to the selected travel mode. The steering control unit 124 suspends the lane change and returns to the standard lane keeping mode when the risk expected up to the route point where the lane change ends on this travel route becomes equal to or greater than the threshold value Rthcoll.

In addition, when the selection unit 123 selects the standard left lane change mode or the standard right lane change mode while traveling in the standard lane keeping mode, the steering control unit 124 switches to the standard lane keeping mode last time. The transition to the standard left lane change mode or the standard right lane change mode may not be performed until the reference period elapses. The reference period is, for example, the time required for the vehicle to run stably. In this manner, the steering control unit 124 can prevent the vehicle from becoming unstable due to repeated lane changes in a short period of time.

[Processing procedure for selecting travel route]
FIG. 10 is a flowchart showing a processing procedure for selecting a lane in which a vehicle travels by the automatic driving device 1 of the first embodiment. This processing procedure is started, for example, while the vehicle is traveling by the automatic steering mechanism without the driver holding the steering wheel.

First, the identification unit 121 identifies a travel route along which the vehicle travels in the lane (S101). At this time, the identifying unit 121 identifies collision risks at a plurality of route points on the travel route (S102). The calculation unit 122 obtains the maximum collision risk value R _max ^coll at a plurality of route points on the travel route (S103). The calculation unit 122 calculates the function value C ^path (I) of the cost function using the above formulas (2) and (3) (S104).

The selection unit 123 determines whether or not the function values C ^path (I) corresponding to all driving modes have been calculated (S105). When the selection unit 123 determines that the function value C ^path (I) has been calculated for all driving modes (YES in S105), it determines whether or not the lane change included in the driving route is effective. . Selecting unit 123 selects a driving mode that does not involve a lane change and a driving mode that involves lane change that is determined to be valid, in which the function value C ^path (I) of the cost function calculated by calculating unit 122 is the smallest. A driving mode is selected (S106), and the process ends. When the selection unit 123 determines in the determination of S105 that the function value C ^path (I) corresponding to some driving modes has not been calculated (NO in S105), the process returns to S101.

[Effects of the automatic driving device of the first embodiment]
Calculation unit 122 calculates function value C ^path (I) using a cost function including, as a parameter, constant value C ^mode (I) representing difficulty of transition to a state in which the vehicle travels on the travel route, and selection unit 123 selects one of the lanes based on the function value C ^path (I) calculated by the calculator 122 . The calculation unit 122 and the selection unit 123 can reduce the frequency of lane changes by adjusting the constant value C ^mode (I). Therefore, the calculation unit 122 and the selection unit 123 can prevent the frequency of lane changes from becoming too high.

[Second embodiment]
When selecting a driving mode, the control device 10 of the first embodiment specifies, by long-term prediction, the driving route when the vehicle is driven in each of all selectable driving modes. At this time, there is a problem that the control device 10 repeats the long-term prediction and the like, and the calculation load increases, resulting in an excessive calculation time.

The control device 50 of the second embodiment selects some of the plurality of driving modes as candidates for the driving mode of the vehicle, and shifts the vehicle from the current driving mode to the driving mode selected as a candidate. It is determined whether or not the vehicle satisfies a transition condition for transitioning to . At this time, the control device 50 does not need to specify a driving route or the like corresponding to a driving mode that is not selected as a candidate by long-term prediction. , it is possible to suppress the calculation time from becoming excessive.

FIG. 11 is a diagram showing the configuration of the control device 50 installed in the automatic driving device of the second embodiment. The control device 50 includes a storage section 51 and a control section 52 . The storage unit 51 is the same as the storage unit 11 in FIG. The control unit 52 is, for example, a CPU. By executing the programs stored in the storage unit 51, the control unit 52 functions as an acquisition unit 521, a prediction unit 522, an operation identification unit 523, a mode identification unit 524, a candidate selection unit 525, a determination unit 526, and a transition unit 527. Function.

[Acquisition of status information and environment information]
Acquisition unit 521 acquires state information indicating the state of the vehicle. For example, the acquisition unit 521 acquires the detection result as to whether or not the direction indicator is blinking from the vehicle detection unit 2 (FIG. 1). Acquisition unit 521 acquires environment information indicating the state of the surrounding environment of the vehicle. The state of the surrounding environment is, for example, the position, moving direction, moving speed, acceleration, etc. of objects (eg, other vehicles, bicycles, pedestrians, etc.) around the vehicle. Acquisition unit 521 outputs the acquired state information to determination unit 526 . Acquisition unit 521 outputs the acquired environment information to prediction unit 522 .

Acquisition unit 521 acquires environmental information including information indicating traffic conditions around the vehicle. For example, the traffic conditions around the vehicle include, for example, whether there are other lanes on the left and right of the vehicle, whether another vehicle is running in front of the vehicle, and the like. The acquisition unit 521 acquires environment information including information indicating traffic conditions from the environment recognition unit 4 or the map database 6 (FIG. 1). Acquisition unit 521 outputs information indicating traffic conditions included in the acquired environment information to candidate selection unit 525 .

The prediction unit 522 predicts the future state of the surrounding environment based on the environmental information acquired by the acquisition unit 521 . For example, the prediction unit 522 predicts the position of the object in several seconds based on the position of the surrounding object acquired by the acquisition unit 521, the moving direction of the object, and the moving speed of the object. The prediction unit 522 outputs information indicating the predicted future state of the surrounding environment to the determination unit 526 .

The operation identifying unit 523 identifies the operation of the vehicle driver. For example, the operation identifying unit 523 identifies a driver's operation to turn on/off the turn signal lights or a driver's operation to increase or decrease the target vehicle speed. Operation identifying section 523 outputs information indicating the identified operation to candidate selecting section 525 .

The mode identification unit 524 identifies the current driving mode of the vehicle from among a plurality of driving modes indicating a driving plan in automatic driving of the vehicle. FIG. 12 is a diagram showing an example of multiple running modes. In the example of FIG. 12, the vehicle travels in either a standard left lane change mode, a standard lane keeping mode, a standard right lane change mode, a fast lane keeping mode, or a slow lane keeping mode.

In FIG. 12, the standard left lane change mode, standard lane maintenance mode, and standard right lane change mode are the same as in the example of FIG. The high speed lane keeping mode is a driving mode in which the current lane is driven at a higher target speed than the standard lane keeping mode. The low-speed lane keeping mode is a driving mode in which the current lane is driven at a target speed lower than that in the standard lane keeping mode. The target speed for each driving mode shall be specified in advance by the driver or the automatic driving system. Mode identifying portion 524 identifies the current driving mode of the vehicle and outputs information indicating the identified driving mode to candidate selecting portion 525 .

[Selection of driving mode candidates]
Candidate selection unit 525 selects some of the driving modes other than the driving modes specified by mode specifying unit 524 as candidates for the running mode of the vehicle. In the example of this specification, the candidate selection unit 525 selects only one driving mode from among the plurality of driving modes as a candidate for the driving mode of the vehicle, but may select a plurality of driving modes as candidates.

FIG. 13 is a diagram showing an example of selection of driving mode candidates by candidate selection unit 525 . Candidate selection unit 525 selects, as a candidate, one of a plurality of driving modes corresponding to a combination of a plurality of lanes on which the vehicle travels and a plurality of target speeds. For example, if the current running mode of the vehicle is the standard lane keeping mode, the candidate selection unit 525 selects one of the four running modes other than the standard lane keeping mode as a running mode candidate. At this time, the candidate selection unit 525 selects a new running mode candidate every predetermined time. In the example of FIG. 13, the candidate selection unit 525 selects the next driving mode every predetermined time in the order of high speed lane keeping mode, standard left lane change mode, low speed lane keeping mode, standard right lane change mode, and high speed lane keeping mode. Select as candidate. The predetermined time is, for example, the time assumed to be required for determining whether the vehicle satisfies a transition condition for transitioning the vehicle from the current driving mode of the vehicle to the driving mode selected as a candidate.

[Selection of candidates based on traffic conditions]
Further, the candidate selection unit 525 assigns priority to a plurality of driving modes other than the driving mode specified by the mode specifying unit 524 based on traffic conditions, and selects a part of the driving modes based on the assigned priority. may be selected as candidates. For example, when the acquisition unit 521 acquires a traffic condition indicating that there is another lane on the right side of the current lane and no other lane on the left side of the current lane, the candidate selection unit 525 selects assign a relatively low priority to driving modes with lane changes. At this time, candidate selection unit 525 assigns a relatively high priority to the driving mode that involves a lane change to the adjacent lane on the right.

Candidate selection unit 525 increases the frequency of selecting a driving mode with a higher assigned priority as a candidate, and decreases the frequency of selecting a driving mode with a lower assigned priority as a candidate. For example, the candidate selection unit 525 selects a driving mode with a higher assigned priority as a candidate at a frequency of once every four times, and selects a driving mode with a lower assigned priority as a candidate at a frequency of once every eight times. to select. In this way, the candidate selection unit 525 can easily select the driving mode suitable for the traffic condition acquired by the acquisition unit 521 as a candidate. In addition, when the acquisition unit 521 acquires a traffic condition indicating that another vehicle is traveling in front of the vehicle, the candidate selection unit 525 assigns a relatively high priority to the traveling mode that involves a lane change, A relatively low priority may be assigned to driving modes that do not involve lane changes.

[Selection of candidates based on driver's operation]
Candidate selection unit 525 may select some driving modes as candidates based on the operation specified by operation specifying unit 523 . For example, the candidate selection unit 525 may select the standard left lane change mode as a candidate when the operation identification unit 523 identifies the driver's operation to turn on the left turn signal. When the driver turns on the left direction indicator, it can be estimated that the driver intends to change lanes to the adjacent lane on the left. At this time, the candidate selection unit 525 can select the lane that matches the driver's intention as a candidate by selecting the lane change mode that accompanies the lane change to the adjacent lane on the left.

Further, when the operation identifying unit 523 identifies an operation by the driver to increase the target vehicle speed, the candidate selecting unit 525 selects a driving mode (for example, high-speed lane keeping mode) corresponding to a relatively high target vehicle speed as a candidate. may When the driver performs an operation to increase the target vehicle speed, it can be estimated that the driver intends to increase the target vehicle speed. At this time, candidate selection unit 525 can select a candidate driving mode that matches the intention of the driver by selecting a driving mode corresponding to a relatively high target vehicle speed as a candidate.

[Calculation of function value of cost function]
Based on the state information acquired by the acquisition unit 521 and the state of the future surrounding environment predicted by the prediction unit 522, the determination unit 526 selects the driving mode selected by the candidate selection unit 525 from the driving modes specified by the mode specification unit 524. It is determined whether or not the vehicle satisfies a transition condition for transitioning the vehicle to the driving mode. First, the determination unit 526 estimates the collision risk when the vehicle travels in the travel mode selected by the candidate selection unit based on the future surrounding environment state predicted by the prediction unit 522 .

More specifically, based on the state of the future surrounding environment predicted by the prediction unit 522, the determination unit 526 selects a driving route corresponding to the driving mode selected by the candidate selection unit 525 in the same manner as the identification unit 121 in FIG. identify. At this time, the determination unit 526 identifies collision risks at a plurality of route points on the identified travel route. Similarly, the determination unit 526 identifies the travel route corresponding to the current travel mode identified by the mode identification unit 524 and the collision risks at a plurality of route points on the travel route.

The determination unit 526 obtains the maximum values R _max ^coll of a plurality of collision risks at a plurality of route points on the travel route corresponding to the travel mode selected by the candidate selection unit 525 . The determination unit 526 obtains the _distance st end of the travel route along which the vehicle travels in the travel mode selected by the candidate selection unit 525 . Similarly, for the travel route corresponding to the current travel mode specified by the mode specifying unit 524, the determination unit 526 determines the maximum collision risk value R _max ^colll and the length of the travel route _st end and

A determination unit 526 determines whether the vehicle satisfies the transition condition based on the estimated collision risk or road length _stend . For example, the determining unit 526 calculates the function value C ^path (I) of the cost function from equations (2) and (3) using the calculated maximum collision risk value R _max ^coll and the length of the road st _tend . calculate. Determining unit 526 similarly calculates function value C ^path (I) corresponding to the current running mode specified by mode specifying unit 524 .

Determination unit 526 determines whether or not the lane change included in the travel route involving the lane change is valid. For example, when the travel route satisfies Equations (4) and (6), the determination unit 526 determines that the lane change included in this travel route is valid.

Based on the state information acquired by the acquisition unit 521, the determination unit 526 identifies the state of the turn signal lamp of the vehicle. Determination unit 526 determines whether or not the vehicle turn signal lamps satisfy the conditions corresponding to the driving mode selected by candidate selection unit 525 . For example, when the candidate selection unit 525 selects a driving mode that involves a lane change to the adjacent lane on the left, the left direction indicator light of the vehicle is blinking, and the direction indicator light starts blinking. It is determined whether or not the direction indicator satisfies the condition that a predetermined time has elapsed since the start of the turn signal.

[Judgment of transition condition]
Determination unit 526 determines that the lane change included in the travel route corresponding to the travel mode selected by candidate selection unit 525 is valid, and determines that the function value C ^path (I) corresponding to the selected travel mode corresponds to the current travel mode. It is determined that the vehicle satisfies the transition condition when the turn signal lamp satisfies the condition corresponding to the selected driving mode, which is lower than the function value C ^path (I) corresponding to the mode. On the other hand, when determination unit 526 determines that the lane change included in the driving mode selected by candidate selection unit 525 is not valid, determination unit 526 determines that the vehicle does not satisfy the transition condition.

If the function value C ^path (I) corresponding to the driving mode selected by the candidate selecting part 525 is greater than or equal to the function value C ^path (I) corresponding to the current driving mode, the determination part 526 determines that the vehicle satisfies the transition condition is not satisfied. If determining unit 526 determines that the turn signal lamp does not satisfy the condition corresponding to the driving mode selected by candidate selecting unit 525, it determines that the vehicle does not satisfy the transition condition. Determination section 526 outputs to transition section 527 information indicating the determination result as to whether or not the transition condition is satisfied.

[Transition of driving mode]
Transition unit 527 transitions the vehicle to a driving mode selected by candidate selection unit 525 from among a plurality of driving modes. The transition unit 527 transitions the vehicle to the driving mode selected by the candidate selection unit 525 when the determination unit 526 determines that the vehicle satisfies the transition condition. On the other hand, transition unit 527 does not transition the vehicle to the driving mode selected by candidate selection unit 525 when determination unit 526 determines that the vehicle does not satisfy the transition condition.

Further, the transition unit 527 determines that the lane change included in the travel route corresponding to the travel mode selected by the candidate selection unit 525 is valid, and the function value C ^path (I) corresponding to this travel mode is the current travel route. is lower than the function value C ^path (I) corresponding to the mode, but when it is determined that the turn signal does not satisfy the conditions corresponding to the running mode selected by the candidate selection unit 525, the turn signal corresponding to the running mode may flash. The transition unit 527 may transition the vehicle to the driving mode selected by the candidate selection unit 525 after a predetermined period of time has elapsed since the blinking of the turn signal lights started.

Moreover, the transition unit 527 is not limited to the example of transitioning the vehicle to any one of the five driving modes shown in FIG. 12 . For example, the transition unit 527 may transition the vehicle to a high speed left lane change mode, a low speed left lane change mode, a high speed right lane change mode, a low speed right lane change mode, or the like shown in FIG. In addition, the transition unit 527 switches the vehicle to a driving mode in which the vehicle is changed to a lane two lanes adjacent to the lane in which the vehicle is currently driving, a driving mode in which the vehicle is driven in the current lane at a target vehicle speed higher than that in the high-speed lane maintenance mode, or the like. may be transitioned. The target vehicle speed corresponding to each driving mode shall be specified in advance by the driver or the automatic driving system.

[Procedure for determining whether or not the vehicle satisfies the transition condition by the automatic driving device]
FIG. 14 is a flow chart showing a processing procedure of transition of driving modes by the automatic driving device of the second embodiment. This processing procedure is started, for example, while the vehicle is traveling by an automatic steering mechanism in which the driver does not grip the steering wheel.

First, the acquisition unit 521 acquires state information indicating the state of the vehicle and environment information indicating the state of the surrounding environment of the vehicle (S201). The mode identification unit 524 identifies the current running mode of the vehicle (S202). The candidate selection unit 525 selects one driving mode as a candidate for the driving mode of the vehicle from among a plurality of driving modes other than the driving mode specified by the mode specifying unit 524 (S203). The prediction unit 522 predicts the future state of the surrounding environment based on the environment information acquired by the acquisition unit 521 (S204).

The determination unit 526 determines whether the vehicle satisfies a transition condition for transitioning the vehicle from the driving mode identified by the mode identification unit 524 to the driving mode selected by the candidate selection unit 525 (S205). When the determination unit 526 determines that the vehicle satisfies the transition condition (YES in S205), the transition unit 527 transitions the vehicle to the driving mode selected by the candidate selection unit 525 (S206). The operation identifying unit 523 determines whether or not the driver's operation for ending the automatic operation function and switching to manual operation has been identified (S207). The transition unit 527 ends the process when the candidate selection unit 525 identifies the driver's operation for switching to manual operation (YES in S207).

When the determination unit 526 determines that the vehicle does not satisfy the transition condition in the determination of S205 (NO in S205), the transition unit 527 returns to the process of S201. If the operation specifying unit 523 does not specify the driver's operation for switching to manual driving in the determination of S207 (NO in S207), the operation specifying unit 523 returns to the process of S201.

[Effects of the automatic driving device according to the second embodiment]
Determination unit 526 determines whether the vehicle satisfies the transition condition only between the driving mode specified by mode specifying unit 524 and the running mode selected by candidate selection unit 525 . At this time, the determination unit 526 does not need to specify the travel route or the like corresponding to the travel mode not selected by the candidate selection unit 525 by long-term prediction. Calculation load can be reduced. Therefore, determination unit 526 can suppress transition to an erroneous driving mode due to delay in lane change determination.

Although the present invention has been described above using the embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments, and various modifications and changes are possible within the scope of the gist thereof. be. For example, all or part of the device can be functionally or physically distributed and integrated in arbitrary units. In addition, new embodiments resulting from arbitrary combinations of multiple embodiments are also included in the embodiments of the present invention. The effect of the new embodiment caused by the combination has the effect of the original embodiment.

1 Automatic driving device 2 Vehicle detection unit 4 Environment recognition unit 6 Map database 10 Control device 11 Storage unit 12 Control unit 50 Control device 51 Storage unit 52 Control unit 121 Identification unit 122 Calculation unit 123 Selection unit 124 Steering control unit 521 Acquisition unit 522 Prediction unit 523 Operation identification unit 524 Mode identification unit 525 Candidate selection unit 526 Judgment unit 527 Transition unit

Claims (7)

an acquisition unit that acquires state information indicating the state of the vehicle and environment information indicating the state of the surrounding environment of the vehicle;
A mode identification unit that identifies the current driving mode of the vehicle from among a plurality of driving modes that indicate a driving plan in automatic driving of the vehicle;
a candidate selection unit that selects some of the plurality of driving modes other than the driving modes identified by the mode identifying unit as candidates for the driving mode of the vehicle;
a prediction unit that predicts a future state of the surrounding environment based on the environmental information;
Transitioning the vehicle from the driving mode identified by the mode identification unit to the driving mode selected by the candidate selection unit based on the state information and the state of the future surrounding environment predicted by the prediction unit. a determination unit that determines whether the vehicle satisfies a transition condition for causing the
a transition unit that transitions the vehicle to the driving mode selected by the candidate selection unit when the determination unit determines that the vehicle satisfies the transition condition;
An automatic driving device. The determination unit estimates a collision risk when the vehicle travels in the driving mode selected by the candidate selection unit based on the state of the future surrounding environment predicted by the prediction unit, and determines the collision risk. Determining whether the vehicle satisfies the transition condition based on
The automatic driving device according to claim 1. The determination unit estimates the length of a travel route traveled by the vehicle in the travel mode selected by the candidate selection unit, and determines whether the vehicle satisfies the transition condition based on the estimated length of the travel route. determine whether or not
The automatic driving device according to claim 1 or 2. The acquisition unit acquires the environment information including information indicating traffic conditions around the vehicle,
The candidate selection unit assigns priorities to the plurality of driving modes other than the driving mode identified by the mode identification unit based on the traffic conditions, and based on the assigned priorities, selects the part Select the driving mode of as a candidate,
The automatic driving device according to any one of claims 1 to 3. The transition unit transitions the vehicle to the driving mode selected by the candidate selection unit from among the plurality of driving modes corresponding to the combinations of the plurality of lanes and the plurality of target speeds.
The automatic driving device according to any one of claims 1 to 4. further comprising an operation identifying unit that identifies an operation of the driver of the vehicle;
The candidate selection unit selects some of the driving modes as the candidates based on the operation identified by the operation identification unit.
The automatic driving device according to any one of claims 1 to 5. The candidate selection unit selects a new candidate for the driving mode every predetermined time.
The automatic driving device according to any one of claims 1 to 6.

Images