Electronic control device and trajectory generation method

Abstract

According to the present invention, a vehicle-mounted electronic control device has: a risk generating unit that determines a travel risk degree in the periphery of a vehicle on the basis of exterior information acquired by an exterior sensor installed on the vehicle, motion information of the vehicle, and map information, and generates a risk map that maps the travel risk degree on a map; and a track generating unit that generates a travel track of the vehicle using the risk map. The track generating unit generates at least one binarized image in which the risk map is divided into a low risk region and a high risk region by means of a binarization process for determining whether the travel risk degree is greater than a predetermined setting value, performs a thinning process on the at least one binarized image, determines a center line in the width direction in the low risk region, and generates the travel track of the vehicle on the basis of the determined center line.

Description

[Technical Field]

[0001] The present invention relates to an electronic control device mounted on a vehicle, and more particularly to a trajectory generation method. [Background technology]

[0002] In recent years, in order to realize comfortable and safe driving assistance and autonomous driving of vehicles, electronic control devices have been proposed that calculate the driving risk of the vehicle based on the surrounding environment of the vehicle recognized by sensors and provide driving assistance according to the calculated risk.

[0003] The following prior art is included as background art in this technical field: Patent Document 1 (JP 2017-41149 A) ​​describes a path generation device including: a cell division unit that generates cells formed by dividing a plane into a grid of a predetermined size based on floor plan data including information representing the types of components of a house and sets information representing the type of the component in each of the cells; a thinning unit that reads the information representing the type of the component set in the cell, identifies the outer edge of a passable area based on whether or not the component can be passed through in advance and associated with the type of the component, and generates path data by thinning the passable area; a graph generation unit that generates a graph including nodes representing endpoints or intersections of the path and edges connecting the nodes based on the path data generated by the thinning unit; and a path search unit that sets a start point and an end point based on input from a user and searches for a path from the start point to the end point using the graph generated by the graph generation unit.

[0004] In addition, Patent Document 2 (JP 2020-53069 A) describes an on-board electronic control device that is mounted on a vehicle, and includes: a vehicle information acquisition unit that acquires vehicle information regarding the movement of the vehicle; a presence time range determination unit that determines vehicle presence time range information that represents the presence time range of the vehicle for each position around the vehicle based on the vehicle information; an environmental element presence time range determination unit that determines environmental element presence time range information that represents the presence time range of the environmental element for each position around the vehicle based on the surrounding environmental element information; and a driving risk determination unit that identifies areas around the vehicle where there is a high risk of the vehicle colliding with the environmental element based on the presence time range information and the environmental element presence time range information. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2017-41149 [Patent Document 2] Japanese Patent Application Publication No. 2020-53069 Summary of the Invention [Problem to be solved by the invention]

[0006] When generating a trajectory based on a risk map that maps driving danger levels, a huge amount of calculation is required to search for a trajectory from the risk map that minimizes the sum of evaluation items. Therefore, a method for calculating a trajectory from a risk map with a small amount of calculation is needed. [Means for solving the problem]

[0007] A representative example of the invention disclosed in the present application is as follows: That is, an electronic control device mounted on a vehicle includes a calculation device that executes predetermined processing, and a storage device accessible by the calculation device, and includes a risk map generation unit that determines a driving risk level around the vehicle based on external environment information acquired by an external environment sensor installed on the vehicle, motion information of the vehicle, and map information, and maps the driving risk level on a map to generate a risk map, and a trajectory generation unit that generates a driving trajectory of the vehicle using the risk map, wherein the trajectory generation unit generates at least one binarized image in which the risk map is divided into a low-risk area and a high-risk area by a binarization process that determines whether the driving risk level is higher than a predetermined set value, and performing a thinning process to reduce the width of the vehicle's travel path in the low-risk area; The method is characterized in that a center line in the width direction is determined, and a running trajectory of the vehicle is generated based on the determined center line. [Effects of the Invention]

[0008] According to one aspect of the present invention, a trajectory can be calculated from a risk map with a small amount of calculation. Problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram showing the physical configuration of an electronic control device according to an embodiment of the present invention; [Figure 2] FIG. 2 is a logical configuration diagram of an electronic control device according to an embodiment of the present invention. [Figure 3] FIG. 2 is a logical configuration diagram of a risk map generation unit. [Figure 4A] FIG. 10 is a diagram illustrating a process for generating a risk map. [Figure 4B] FIG. 10 is a diagram illustrating a process for generating a risk map. [Figure 4C] FIG. 10 is a diagram illustrating a process for generating a risk map. [Figure 4D] FIG. 10 is a diagram illustrating a process for generating a risk map. [Figure 4E] FIG. 10 is a diagram illustrating a process for generating a risk map. [Figure 4F] FIG. 10 is a diagram illustrating a process for generating a risk map. [Figure 4G] FIG. 10 is a diagram illustrating a process for generating a risk map. [Figure 4H] FIG. 10 is a diagram illustrating a process for generating a risk map. [Figure 4I] FIG. 10 is a diagram illustrating a process for generating a risk map. [Figure 4J] FIG. 10 is a diagram illustrating a process for generating a risk map. [Figure 4K] FIG. 10 is a diagram illustrating a process for generating a risk map. [Figure 4L] FIG. 10 is a diagram illustrating a process for generating a risk map. [Figure 5] FIG. 10 is a diagram illustrating an example of a risk table in which risk values ​​are defined. [Figure 6] FIG. 2 is a logical configuration diagram of a trajectory generation unit. [Figure 7] FIG. 10 is a diagram showing a binarized risk map. DETAILED DESCRIPTION OF THE INVENTION

[0010] FIG. 1 is a diagram showing the physical configuration of an electronic control unit 1 according to an embodiment of the present invention.

[0011] The electronic control unit (ECU) 1 includes a CPU 11, a nonvolatile memory (ROM: Read Only Memory) 12, a volatile memory (RAM: Random Access Memory) 13, and an accelerator 14. The CPU 11 is a calculation device that executes programs stored in the volatile memory 13. The nonvolatile memory 12 is a storage device with a nonvolatile storage area that retains stored data even when power is cut off, and includes a program area that stores programs executed by the CPU 11 and a data area that temporarily stores data used by the CPU 11 when executing the programs. The volatile memory 13 is a storage device with a volatile storage area that stores data used by the CPU 11 when executing the programs. The accelerator 14 is a calculation device that can process specific calculations at high speed and executes image processing in a risk map generation unit 16 and a trajectory generation unit 17. The electronic control unit 1 communicates with other electronic control units and sensors via networks such as a CAN or Ethernet.

[0012] The electronic control unit 1 receives as input observation results from sensors such as a camera 21, radar 22, and LiDAR 23. The electronic control unit 1 is also connected to a map unit 24 that provides map information using positioning information from a GNSS (Global Navigation Satellite System) device 25. The electronic control unit 1 also receives as input vehicle information 26 that indicates the behavior of the vehicle, such as speed and acceleration. The electronic control unit 1 outputs a drive control signal 31, a braking control signal 32, and a steering control signal 33 to control the running of the vehicle.

[0013] FIG. 2 is a logical configuration diagram of the electronic control unit 1 according to the embodiment of the present invention.

[0014] The electronic control device 1 has a fusion unit 15, a risk map generation unit 16, a trajectory generation unit 17, and a trajectory tracking unit 18. The fusion unit 15 integrates the observation results of multiple sensors (camera 21, radar 22, LiDAR 23, etc.) to identify the position, size, and type of targets around the vehicle. The risk map generation unit 16 generates a risk map indicating the driving risk around the vehicle based on the identified targets. The trajectory generation unit 17 uses the generated risk map to generate a trajectory for the vehicle to travel. The trajectory tracking unit 18 generates a drive control signal 31, a braking control signal 32, and a steering control signal 33 to control the vehicle according to the generated trajectory. The drive control signal 31 is a signal for controlling the rotation of the drive source (engine, motor), i.e., the acceleration and speed. The braking control signal 32 is a signal for controlling deceleration by braking. The steering control signal 33 is a signal for controlling the direction of travel of the vehicle by steering.

[0015] FIG. 3 is a logical configuration diagram of the risk map generating unit 16. As shown in FIG.

[0016] The risk map generation unit 16 first initializes various parameters (61). For example, the size of the grid is determined based on the minimum size of a target object that is a risk target and the detection range of the sensor, and the image plane 101 is defined. More specifically, if the width of the white lines that demarcate road boundaries and lanes is 10 cm and the detection range is 300 m square, the grid size is set to 10 cm, which is approximately the same as the width of the white lines, and a risk map of 3000 × 3000 pixels (see FIG. 4A) is created within an area 300 m ahead of the host vehicle 100 and 150 m to the left and right, and a memory area for storing the image plane 101 of this risk map is reserved.

[0017] Next, the risk map generation unit 16 calculates the movement range from the road structure (63). For example, the movement range of the vehicle is calculated based on the road structure in the traveling direction of the vehicle (e.g., road curves, intersections), and movement range 102 within the specified image plane 101 is extracted from the calculated movement range. For example, as shown in FIG. 4B, if the road curves to the right, the area to the left of the road is excluded, and the area to the right is extracted as movement range 102. The road structure is calculated from map data acquired from the map unit 24 or observation results of landmarks (e.g., road boundaries) observed by a sensor.

[0018] Next, the risk map generating unit 16 limits the movement range ahead of the vehicle (64). For example, the movement range is limited to the range that the vehicle can travel within a predetermined time depending on the traveling speed of the vehicle. Specifically, a rectangle 103 is defined ahead of the vehicle (FIG. 4C). It is preferable that the rectangle 103 is short in the traveling direction at low speeds and long in the traveling direction at high speeds.

[0019] Next, the risk map generation unit 16 calculates the apparent risk of stationary obstacles (65). For example, since the presence or absence of an obstacle is unknown outside the sensor's detection range 104, the area is mapped as a high-risk area (see FIG. 4D). The triangular high-risk area 104 prevents discontinuity between the vehicle position and the generated trajectory, enabling the generation of a smooth trajectory on which the vehicle can travel. Furthermore, the risk map generation unit 16 maps stationary obstacles observed by the sensor and stationary obstacles included in the map data acquired from the map unit 24 as high-risk areas based on their positions, sizes, and orientations on the risk map. At this time, the risk table (FIG. 5) is referenced to determine a different driving risk level for each stationary obstacle. Note that it is preferable to map high-risk areas around stationary obstacles according to the sensor's observation error. For example, as shown in FIG. 4E, a stationary obstacle 105 observed by the sensor and its surrounding area 106 are mapped as high-risk areas on the risk map. The stationary obstacle area 105 and the surrounding area 106 may be assigned different driving risk levels. Furthermore, high-risk areas may be mapped in the surrounding area 106 of the stationary obstacle 105 depending on the type observed (for example, the area where the door of a parked vehicle opens).

[0020] The risk map generating unit 16 updates the previous value of the apparent risk of the stationary obstacle, which is input in step 63, with the calculated value (62).

[0021] Next, the risk map generation unit 16 calculates the potential risk of stationary obstacles (66). For example, the presence or absence of a blind spot is determined based on the stationary obstacles observed by the sensor and the height of the stationary obstacles included in the map data acquired from the map unit 24. If a blind spot is determined to exist, a potential risk area 107 is mapped on the risk map based on the location and type of the blind spot. At this time, the risk table (FIG. 5) is referenced to determine a different driving risk level for each stationary obstacle. Specifically, in the case shown in FIG. 4F, a potential risk area 107 where a pedestrian may jump out from the shadow of a stopped vehicle 106 is mapped.

[0022] Next, the risk map generation unit 16 restricts the movement range from the current position of the vehicle (67). For example, the movement range is restricted to a range within which the vehicle can move at the current speed of the vehicle in accordance with a lateral acceleration constraint (for example, 0.1 G or less) derived from the vehicle's driving performance, and is mapped on the risk map as a high-risk area 108 (see FIG. 4G). At this time, the risk table (FIG. 5) is referenced to determine the driving risk level.

[0023] Next, the risk map generation unit 16 calculates the actual risk of the moving obstacle (68). For example, the moving obstacle observed by the sensor is mapped on the risk map as a high-risk area based on its position, size, and orientation. At this time, the risk table (FIG. 5) is referenced to determine the driving risk level. Note that the risk map generation unit 16 may map the area of ​​the moving obstacle on the predicted path as a high-risk area, taking into account the speed change and predicted path of the moving obstacle. For example, as shown in FIG. 4H, the moving obstacle 109 observed by the sensor and its predicted area 110 are mapped on the risk map as high-risk areas. At this time, the current moving obstacle area 109 and the predicted moving obstacle area 110 may be assigned different driving risk levels. Also, the risk of the current location may be lowered, and the risk of the position on the predicted path may be higher.

[0024] Next, the risk map generation unit 16 calculates the potential risk of the moving obstacle (69). For example, the presence or absence of a blind spot is determined based on the height of the moving obstacle observed by the sensor. If a blind spot is determined to exist, a potential risk area 107 is mapped based on the location and type of the blind spot.

[0025] Next, the risk map generation unit 16 restricts the dynamic movement range of the host vehicle (70). For example, as shown in FIGS. 4I to 4K, an unavoidable area 111 is calculated from driving ranges 112a and 112b of the host vehicle in which the obstacle can be avoided in each direction due to the constraint of the lateral acceleration (for example, 0.1 G or less) occurring at the current speed of the host vehicle (the overlapping area of ​​112a and 112b that corresponds to the vehicle width of the preceding vehicle), and is mapped on the risk map as a high-risk area 111. At this time, the risk table (FIG. 5) is referenced to determine the driving risk level. Specifically, the driving range of the host vehicle in which the moving obstacle can be avoided is calculated taking into account the lateral acceleration limit value, and the area in which the moving obstacle cannot be avoided is mapped on the risk map as a high-risk area between the moving obstacle and the host vehicle, as shown in FIG. 4L.

[0026] FIG. 5 is a diagram showing an example of a risk table in which risk values ​​are defined.

[0027] The risk table defines risk values ​​based on road structure (color, shape, position, and meaning of road surface paint, road shape, type and height of sidewalls), obstacles (type of obstacle and whether it is moving), type of blind spot, etc. The risk table may define different risk values ​​based on the external information shown in the figure, as well as vehicle behavior (vehicle speed, acceleration, etc.), driving environment, and driver operation (turn signal operation, etc.). The risk table is referenced by the risk map generation unit 16, which calculates the risk of obstacles observed or on a map. The risk table defines risk values ​​that are higher for obstacles with a higher driving risk.

[0028] FIG. 6 is a logical configuration diagram of the trajectory generation unit 17.

[0029] First, the trajectory generation unit 17 determines a threshold value to be used in the binarization process (71). For example, the threshold value to be used in the binarization process may be changed depending on the situation around the vehicle. Furthermore, this threshold value may be changed depending on the situation around the vehicle (external information), as well as vehicle behavior (vehicle speed, acceleration, etc.), the driving environment, and driver operation (turn signal operation, etc.). More specifically, when many targets are observed around the vehicle, such as when the area around the vehicle is congested, the threshold value may be increased to generate a safer trajectory. When few targets are observed around the vehicle, the threshold value may be decreased to increase the number of trajectory options. Furthermore, multiple trajectories with different driving risk levels may be generated using multiple threshold values, and a trajectory may be selected in a later process. For example, the threshold value may be set to 15 to increase the degree of latitude in complying with traffic rules, and to 55 to decrease the degree of latitude in complying with traffic rules and prioritize minimizing accidents.

[0030] Next, the trajectory generating unit 17 executes binarization processing (7 2 For example, the binarization process assigns a value of 0 to areas where the driving risk is lower than the threshold determined in step 71, and a value of 1 to areas where the driving risk is higher than the threshold. The risk map is then visualized in binary form by coloring low-risk areas with a value of 0 in black and high-risk areas with a value of 1 in white (see Figure 7).

[0031] Next, the trajectory generating unit 17 executes an expansion process to expand the white area where the driving risk is high (7 3 ) For example, the expansion process is performed the number of times that the decimal point of the value obtained by dividing half the vehicle width by the grid size determined in the initialization process is rounded up. The expansion process can remove noise from the black low-risk areas. In other words, the areas in which the vehicle can travel while avoiding driving hazards are determined as black low-risk areas. After that, the black and white are reversed for the next process, and the low-risk areas are colored white and the high-risk areas are colored black.

[0032] Next, the trajectory generating unit 17 executes thinning processing (7 4This thinning process generates lines with a line width of one pixel, determines the center line of the width of the low-risk area, and generates the determined center line as the vehicle trajectory.

[0033] Next, the trajectory generation unit 17 acquires a risk value on the trajectory generated by the thinning process (75). Then, the trajectory generation unit 17 calculates the vehicle's traveling speed from the acquired risk value (76). For example, the target vehicle speed vt is calculated using the following formula, where vkm / h is the lower of the current vehicle speed or the speed limit, and x% is the risk value. vt=v×(1.0-x / 100)

[0034] If the trajectory ends midway, the target vehicle speed vt at the end of the trajectory is set to 0, and the target vehicle speed is calculated so that the vehicle can be stopped smoothly without sudden deceleration. However, the target vehicle speed vt is not set to 0 at the point where the trajectory ends at the end of the risk map.

[0035] As described above, the electronic control device according to the embodiment of the present invention can calculate a trajectory from a risk map through simple repetitive processing. This makes it ideal for computing devices such as accelerators, and can shorten the processing time for trajectory generation. In particular, since the risk map can be created through image processing such as patterning and overlay, the processing time for creating the risk map can be shortened. Furthermore, by taking into account unavoidable areas when creating the risk map, a smooth trajectory can be created when performing avoidance operations to the left or right.

[0036] The present invention is not limited to the above-described embodiments, but includes various modifications and equivalent configurations within the spirit and scope of the appended claims. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to configurations including all of the described configurations. Furthermore, part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Furthermore, the configuration of another embodiment may be added to the configuration of one embodiment. Furthermore, part of the configuration of each embodiment may be added, deleted, or replaced with other configurations.

[0037] Furthermore, the aforementioned configurations, functions, processing units, processing means, etc. may be realized in part or in whole in hardware, for example by designing them as integrated circuits, or may be realized in software by having a processor interpret and execute a program that realizes each function.

[0038] Information such as programs, tables, and files that realize each function can be stored in a storage device such as a memory, a hard disk, or an SSD (Solid State Drive), or in a recording medium such as an IC card, an SD card, or a DVD.

[0039] In addition, the control lines and information lines shown are those that are considered necessary for explanation, and do not necessarily represent all the control lines and information lines that are necessary for implementation. In reality, it can be assumed that almost all components are interconnected.

Claims

1. An electronic control device mounted on a vehicle, A computing device that executes predetermined processing and a storage device that can be accessed by the computing device, a risk map generation unit that determines a driving risk level around the vehicle based on external environment information acquired by an external environment sensor installed in the vehicle, motion information of the vehicle, and map information, and maps the driving risk level on a map to generate a risk map; a trajectory generation unit that generates a traveling trajectory of the vehicle using the risk map, The trajectory generation unit generating at least one binarized image in which the risk map is divided into a low risk area and a high risk area by a binarization process that determines whether the driving risk is greater than a predetermined set value; performing a thinning process to reduce a width of the vehicle's travel path in the low-risk area in the at least one binarized image, and determining a center line in the width direction; An electronic control device that generates a running trajectory of the vehicle based on the determined center line.

2. 2. The electronic control device according to claim 1, The trajectory generation unit changes the setting value based on at least one of the external information, vehicle information indicating the behavior of the vehicle including at least speed and acceleration, and the driving environment of the vehicle.

3. 2. The electronic control device according to claim 1, The risk map generation unit determines the driving risk level based on at least one of the external information, vehicle information indicating the behavior of the vehicle including at least speed and acceleration, and the driving environment of the vehicle.

4. 2. The electronic control device according to claim 1, The electronic control device is characterized in that the trajectory generation unit reduces the width of the traveling trajectory by an amount equivalent to the vehicle width in the thinning process.

5. 2. The electronic control device according to claim 1, An electronic control device characterized in that it maps areas where driving is not possible on a map as areas where driving danger is high based on the driving performance of the vehicle.

6. 2. The electronic control device according to claim 1, The electronic control device is characterized in that the trajectory generation unit generates a plurality of travel trajectories using a plurality of the set values.

7. A trajectory generation method executed by an on-vehicle electronic control device, comprising: The on-vehicle electronic control device has a calculation device that executes predetermined processing and a storage device that can be accessed by the calculation device, The trajectory generation method includes: a risk map generation step in which the computing device determines a driving risk level around the vehicle based on external environment information acquired by an external environment sensor installed in the vehicle, motion information of the vehicle, and map information, and generates a risk map by mapping the driving risk level on a map; a trajectory generation step in which the calculation device generates a traveling trajectory of the vehicle using the risk map; In the trajectory generation step, the calculation device generating at least one binarized image in which the risk map is divided into a low-risk area and a high-risk area by a binarization process that determines whether the driving risk is greater than a predetermined set value; performing a thinning process to reduce the width of the vehicle's travel path in the low-risk area in the at least one binarized image, and determining a center line in the width direction; A trajectory generation method, comprising: generating a running trajectory of the vehicle based on the determined center line.

Images