Behavior planning device and control computing device

Abstract

The behavior planning device (102) comprises: a plane coordinate system movement prediction unit (104) for predicting the movement of the obstacle based on plane coordinate system obstacle information (S3) in which the obstacle detected by the external sensor of the vehicle (100) is expressed in the plane coordinate system, and outputting the predicted movement as plane coordinate system obstacle movement information (S7); a scene determination unit (105) for determining the state of the obstacle based on the plane coordinate system obstacle movement information (S7), and outputting the state of the vehicle (100) as scene information (S8); and a behavior determination unit (106) for determining the behavior of the vehicle (100) based on the scene information (S8), and outputting the predicted movement as a behavior determination result (S9). A control operation unit (103) calculates a target value for controlling the vehicle (100) according to the behavior determination result (S9) outputted from the behavior planning device (102).

Description

Technical Field

[0001] The present disclosure relates to a behavior planning device that appropriately determines the behavior of a host vehicle in an automatic driving system, and a control calculation device that calculates a target value for controlling the host vehicle based on the determined behavior. Background Art

[0002] In recent years, the development of automatic driving systems for driving cars has been progressing. In automatic driving systems, the behavior of the vehicle needs to be appropriately determined based on the position and speed of obstacles such as pedestrians, bicycles, and other vehicles around the vehicle. Here, the behavior of the vehicle refers to, for example, maintaining the driving lane, changing the driving lane, and stopping.

[0003] Patent document 1 discloses a vehicle surrounding information verification device that detects obstacles around the vehicle and arranges them on a map, and determines the behavior based on the obstacles. The vehicle surrounding information verification device evaluates the reliability of the obstacle detection result by comparing the position of the obstacle with the drivable area, and selects whether to apply the determined behavior based on the evaluation result. As a result, even if the accuracy of obstacle detection decreases, the wrong behavior will not be determined.

[0004] Non-patent document 1 discloses descriptions related to high-precision maps.

[0005] Prior art literature

[0006] Patent Document 1: WO2016 / 166790

[0007] Non-patent document 1: "Automatic walking robot maintenance", Sesame / Control / Information Publishing, Vol. 60, No. 11, 2016, p. 463-468 Summary of the invention

[0008] In order to determine the behavior of the host vehicle, it is necessary to determine the situation of the host vehicle based on the future movement prediction of the obstacle. In the conventional vehicle surrounding information verification device, there is no mention of determining the situation of the host vehicle, so it is difficult to appropriately determine the behavior of the host vehicle.

[0009] The present disclosure has been made to solve the above-mentioned problems, and an object of the present disclosure is to provide a behavior planning device and a control calculation device that appropriately determine the behavior of a host vehicle and improve the accuracy of autonomous driving.

[0010] The behavior planning device involved in the present disclosure comprises: a plane coordinate system movement prediction unit, which predicts the movement of the obstacle based on plane coordinate system obstacle information that represents the surrounding obstacles detected by an external sensor installed in the vehicle in a plane coordinate system, and outputs the plane coordinate system obstacle movement information; a scene determination unit, which determines the condition of the obstacle based on the plane coordinate system obstacle movement information, and outputs the condition of the vehicle as scene information; and a behavior determination unit, which determines the behavior of the vehicle based on the scene information, and outputs it as a behavior determination result.

[0011] In addition, the control calculation device according to the present disclosure calculates a target value for controlling the vehicle according to the behavior determination result output from the behavior planning device.

[0012] According to the present disclosure, the behavior planning device and the control calculation device determine the situation of the vehicle based on the movement prediction of the obstacle, so the behavior of the vehicle can be appropriately determined, and the accuracy of automatic driving can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a diagram showing an example of a configuration of a vehicle equipped with the behavior planning device and the control calculation device in Embodiments 1 to 4.

[0014] Figure 2 This is a block diagram showing an example of the behavior planning device and the control calculation device in the first embodiment.

[0015] Figure 3 This is a schematic diagram showing an example of a scene in the plane coordinate system in Implementation Example 1.

[0016] Figure 4 This is a schematic diagram showing an example of a finite state machine in a behavior determination unit in Embodiments 1 to 4.

[0017] Figure 5 This is a schematic diagram showing an example of a target trajectory from the motion planning unit in the first to fourth embodiments.

[0018] Figure 6 This is a block diagram showing an example of a behavior planning device and a control calculation device in the second embodiment.

[0019] Figure 7 This is a schematic diagram showing an example of a vehicle in the plane coordinate system and the path coordinate system in Embodiments 2 to 4.

[0020] Figure 8 This is a schematic diagram showing an example of the positional relationship between the vehicle and an obstacle when traveling on a curve in Embodiments 2 to 4.

[0021] Fig. 9 This is a schematic diagram showing an example of the positional relationship between the vehicle and road information when traveling along a curve in Embodiments 2 to 4.

[0022] Fig.10 This is a schematic diagram showing an example of a scene in the path coordinate system in Embodiments 2 to 4.

[0023] Fig.11 This is a block diagram showing an example of a behavior planning device and a control calculation device in the third embodiment.

[0024] Fig.12 This is a schematic diagram showing an example of the positional relationship between the vehicle and an obstacle when traveling at an intersection in the third and fourth embodiments.

[0025] Fig.13 This is a schematic diagram showing an example of the positional relationship between the vehicle and an obstacle when traveling on a T-junction road in the third and fourth embodiments.

[0026] Fig.14 This is a block diagram showing an example of a behavior planning device and a control calculation device in a fourth embodiment. DETAILED DESCRIPTION

[0027] The behavior planning device and the control computing device in the embodiment of the present disclosure are described below with reference to the accompanying drawings. In addition, the vehicle is referred to as "vehicle" and pedestrians, bicycles and other vehicles around the vehicle are collectively referred to as "obstacles".

[0028] Implementation method 1.

[0029] Figure 1 1 is a diagram showing an example of the structure of a vehicle 100 equipped with the behavior planning device 102 and the control operation device 103 in the first embodiment. Figure 1 In the embodiment, the behavior planning device 102 and the control operation device 103 are combined to form the automatic driving system 101.

[0030] A steering wheel 1 provided for a driver (i.e., a driver) to operate the vehicle 100 is combined with a steering shaft 2. A pinion shaft 13 of a rack and pinion mechanism 4 is connected to the steering shaft 2. A rack shaft 14 of the rack and pinion mechanism 4 is freely movable back and forth according to the rotation of the pinion shaft 13, and a front steering knuckle 6 is connected to the left and right ends thereof via a tie rod 5. The front steering knuckle 6 supports a front wheel 15 as a main steering wheel in a freely rotatable manner, and is supported on a vehicle body frame in a freely steerable manner.

[0031] The torque generated by the driver operating the steering wheel 1 rotates the steering shaft 2, and the rack and pinion mechanism 4 moves the rack shaft 14 to the left and right according to the rotation of the steering shaft 2. The front knuckle 6 rotates around a kingpin shaft (not shown) as the center by the movement of the rack shaft 14, thereby turning the front wheel 15 to the left and right. Therefore, the driver can change the lateral displacement of the vehicle 100 by operating the steering wheel 1 when the vehicle 100 moves forward or backward.

[0032] In the vehicle 100, as internal sensors 20 for identifying the driving state of the vehicle 100, a vehicle speed sensor 21, an IMU (Inertial Measurement Unit) sensor 22 for measuring the inertia of the vehicle, a steering angle sensor 23, a steering torque sensor 24, and the like are provided.

[0033] The vehicle speed sensor 21 is mounted on a wheel of the vehicle 100 , includes a pulse sensor for detecting the rotation speed of the wheel, and converts the output of the pulse sensor into a vehicle speed value and outputs the converted value.

[0034] The IMU sensor 22 is disposed on the roof or in the interior of the vehicle 100 to detect the acceleration and angular velocity of the vehicle 100 in the vehicle coordinate system. The IMU sensor 22 also includes sensors such as MEMS (Micro Electric Mechanical System) and fiber optic gyroscopes. Here, the vehicle coordinate system refers to a coordinate system fixed to the chassis or body of the vehicle 100. Usually, in the vehicle coordinate system, the center of gravity of the vehicle 100 is used as the origin, the front of the length direction of the vehicle 100 is defined as the x-axis, the left-hand direction of the vehicle 100 is defined as the y-axis, and the right-handed forward direction starting from the x-axis in the y-axis direction is defined as the z-axis.

[0035] The steering angle sensor 23 is a sensor that measures the rotation angle of the steering shaft 2 , and is composed of, for example, a rotary encoder.

[0036] The steering torque sensor 24 is a sensor that measures the rotation torque of the steering shaft 2 and is composed of, for example, a strain gauge.

[0037] In addition, the vehicle 100 is provided with a camera 25 , a radar 26 , a GNSS (Global Navigation Satellite System) sensor 27 , a LiDAR (Light Detection and Ranging) 29 , and the like as external sensors for recognizing the surrounding conditions of the vehicle 100 .

[0038] The camera 25 is installed at a position capable of photographing the front, sides and rear of the vehicle 100 , and obtains information representing the environment in front of the vehicle 100 , such as lanes, mark lines and obstacles in front of the vehicle 100 , from the photographed images.

[0039] The radar 26 irradiates the front of the vehicle 100 with radar light and detects the reflected wave, thereby measuring the relative distance and relative speed of an obstacle existing in front of the vehicle 100 and outputting the measurement results.

[0040] A GNSS antenna (not shown) is connected to the GNSS sensor 27. The GNSS sensor 27 uses the GNSS antenna to receive positioning signals from positioning satellites orbiting in satellite orbits, analyzes the received positioning signals, and outputs information on the position of the phase center of the GNSS antenna (latitude, longitude, altitude, and azimuth, etc.). Positioning satellites include the United States' GPS (Global Positioning System), Russia's GLONASS (Global Navigation Satellite System), Europe's Galileo, Japan's QZSS (Quasi-Zenith Satellite System), China's BeiDou, and India's NavIC (Navigation Indian Constellation). The GNSS sensor 27 can use any of these positioning satellites.

[0041] LiDAR 29 is installed on the roof of the vehicle 100 or the like. LiDAR 29 detects the position of the object in the vehicle coordinate system by irradiating the periphery of the vehicle 100 with laser light and detecting the time difference until the laser light is reflected by the surrounding objects and returns. In recent years, LiDAR 29 can also detect objects in a wider range and with higher density by installing it at the four corners of the vehicle 100.

[0042] The navigation device 28 internally stores the map information S15, and has the following functions: based on the map information S15, the position information of the vehicle 100 obtained by the GNSS sensor 27, etc., and the destination information set by the driver, it calculates a driving route that can reach the destination and outputs navigation information. The navigation device 28 also has the function of recognizing that the surrounding area of ​​the vehicle 100 is an intersection area and outputting it, and the function of calculating and outputting the instruction of the lane change required to reach the destination.

[0043] The information acquisition unit 30 is connected to external sensors such as the camera 25, the radar 26, and the LiDAR 29, and performs fusion processing on the information acquired by these sensors, thereby detecting information including obstacle information around the vehicle 100, and outputting the information to the automatic driving system 101. The information acquisition unit 30 is also connected to the navigation device 28, and detects the position of the vehicle 100 based on the GNSS sensor 27. The information acquisition unit 30 will be described later. Figure 2 Detailed description.

[0044] The automatic driving system 101 includes a behavior planning device 102 and a control computing device 103. The behavior planning device 102 determines the behavior of the vehicle 100 based on the information from the information acquisition unit 30 and the information from the internal sensor 20, and outputs the behavior determination result S9 to the control computing device 103. The control computing device 103 calculates a target value for controlling the vehicle 100 according to the behavior determination result S9 from the behavior planning device 102. Here, the target value refers to, for example, a target steering amount S11 collectively referred to as a target steering angle and a target steering torque, and a target acceleration / deceleration amount S12.

[0045] The vehicle 100 is provided with an electric motor 3 for realizing lateral movement of the vehicle 100 , a vehicle drive device 7 for controlling movement of the vehicle 100 in the front-rear direction, and actuators such as brakes 11 .

[0046] The electric motor 3 is generally composed of a motor and gears, and can freely rotate the steering shaft 2 by providing torque to the steering shaft 2. That is, the electric motor can freely steer the front wheels 15 independently of the operation of the steering wheel 1 by the driver.

[0047] The steering control device 12 calculates the current value that should be supplied to the electric motor 3 in order to make the steering of the vehicle 100 track the target steering amount S11 based on the output of the steering angle sensor 23, the steering torque sensor 24, etc. and the target steering amount S11 from the automatic driving system 101, and outputs a current equivalent to the calculated current value to the electric motor 3.

[0048] The vehicle drive device 7 is an actuator for driving the vehicle 100 in the front-rear direction. The vehicle drive device 7 rotates the front wheels 15 and the rear wheels 16 via a transmission (not shown) and a drive shaft 8 using a driving force obtained from a driving source such as an engine or a motor. Thus, the vehicle drive device 7 can freely control the driving force of the vehicle 100.

[0049] On the other hand, the brake control device 10 is an actuator for braking the vehicle 100, and controls the braking amount of the brakes 11 respectively provided on the front wheels 15 and the rear wheels 16 of the vehicle 100. The general brake 11 generates a braking force by pressing a brake pad against a disk rotor rotating together with the front wheels 15 and the rear wheels 16 using hydraulic pressure.

[0050] The acceleration / deceleration control device 9 calculates the driving force and braking force of the vehicle 100 required to make the acceleration of the vehicle 100 follow the target acceleration / deceleration amount S12 from the automatic driving system 101 , and outputs the calculation results to the vehicle drive device 7 and the brake control device 10 .

[0051] The internal sensor 20, the external sensor, and the plurality of devices described above form a network using a CAN (Controller Area Network), a LAN (Local Area Network), etc. in the vehicle 100. The devices can obtain various information via the network. In addition, the internal sensor 20 and the external sensor can send and receive data to each other via the network.

[0052] Figure 2 This is a block diagram showing an example of the behavior planning device 102 and the control calculation device 103 in the first embodiment. Figure 2 The block diagram includes an information acquisition unit 30, an internal sensor 20, a behavior planning device 102, a control calculation device 103, a steering control device 12, and an acceleration / deceleration control device 9. The behavior planning device 102 determines the behavior of the vehicle 100 based on the information from the information acquisition unit 30 and the information from the internal sensor 20, and outputs it as a behavior determination result S9. The control calculation device 103 calculates a target value for controlling the vehicle 100 based on the information from the information acquisition unit 30, the information from the internal sensor, and the behavior determination result S9 from the behavior planning device 102. Here, the target value is a target steering amount S11 and a target acceleration / deceleration amount S12.

[0053] The information acquisition unit 30 includes a path detection unit 31 , an obstacle detection unit 32 , a road information detection unit 33 , and a vehicle position detection unit 34 .

[0054] The path detection unit 31 outputs a reference path S1 as a driving reference of the vehicle 100 and a drivable area S2 as an area where the vehicle 100 can travel. The reference path S1 is the center line of the lane identified by detecting the marking line using image data obtained from a camera, etc. The reference path S1 is used as a path provided from the outside in addition to the center line of the lane. For example, in a parking lot, when a path for automatic parking is provided from the outside, the path is used as the reference path S1. The reference path S1 is expressed by a polynomial, a spline curve, etc.

[0055] The drivable area S2 is calculated by fusing information obtained from the camera 25, the radar 26, and the LiDAR 29. For example, in a road with left and right markings, the drivable area S2 is output as an area of ​​the road surrounded by the left and right markings when there is no obstacle. If there is an obstacle, it is output as an area obtained by removing only the obstacle from the area surrounded by the left and right markings.

[0056] The obstacle detection unit 32 outputs the plane coordinate system obstacle information S3. The plane coordinate system obstacle information S3 is obtained by fusing the image data obtained from the camera 25, the radar 26, and the information of the LiDAR 29. The plane coordinate system obstacle information S3 is the position and speed of the obstacle, and the type of the obstacle. The type of obstacle is classified into a vehicle, a pedestrian, a bicycle, a motorcycle, etc. In addition, the position and speed of the obstacle in the plane coordinate system obstacle information S3 are expressed in a plane coordinate system described later. However, it is not limited to the plane coordinate system.

[0057] The road information detection unit 33 outputs road information S4. The road information S4 is a signal device C1 at an intersection and its lighting state detected by fusing the image data obtained from the camera 25, the radar 26, and the LiDAR 29. The road information S4 is not limited to this, and may be a stop line C3 in front of the signal device C1.

[0058] The vehicle position detection unit 34 detects the position of the vehicle 100 based on the GNSS sensor 27 and outputs it as vehicle position information S5. The position of the vehicle 100 from the GNSS sensor 27 is generally expressed in a geographic coordinate system. Regarding the geographic coordinate system, the earth is usually regarded as an ellipsoid, and is expressed by a combination of longitude and latitude representing the horizontal position and the height representing the vertical position on its surface. It is possible to transform an arbitrary point on the geographic coordinate system as a reference point to the NED (North-East-Down) coordinate system, or to transform it to a plane coordinate system through a Gauss-Kruger projection. The NED coordinate system is a coordinate system obtained by taking an arbitrary point expressed in the geographic coordinate system as the origin and taking a coordinate system in the north, east, and vertically upward directions. The plane coordinate system is an XY coordinate system having two axes orthogonal to each other from the origin. The plane coordinate system is used to express markings for identifying road boundaries, the position of the vehicle 100 and obstacles, etc. The plane coordinate system takes, for example, the center of gravity of the vehicle 100 as the origin, the longitudinal direction of the vehicle 100 as the first axis, and the left-hand direction as the second axis. In this case, it is consistent with the vehicle coordinate system. In other examples, it is also possible to take any point on the map as the origin, set the east direction as the first axis, and set the north direction as the second axis. The vehicle position detection unit 34 has the following function: transform the position of the vehicle 100 represented by the geographic coordinate system into the plane coordinate system, and output it as the vehicle position information S5.

[0059] The internal sensor 20 includes a vehicle speed sensor 21 and an IMU sensor 22. The internal sensor is installed in the vehicle 100, and detects the state quantity of the vehicle 100 based on the vehicle speed sensor and the IMU sensor 22, and outputs it as sensor information S6. Figure 1 This has already been explained, so the explanation will be omitted here.

[0060] The behavior planning device 102 includes a plane coordinate system movement prediction unit 104 , a scene determination unit 105 , and a behavior determination unit 106 .

[0061] The plane coordinate system movement prediction unit 104 predicts the movement of the obstacle based on the plane coordinate system obstacle information S3 from the obstacle detection unit 32, and outputs the result as the plane coordinate system obstacle movement information S7. That is, the plane coordinate system movement prediction unit 104 predicts the movement of the obstacle based on the plane coordinate system obstacle information S3 that represents the surrounding obstacles detected by the external sensor installed in the vehicle 100 in a plane coordinate system, and outputs it as the plane coordinate system obstacle movement information S7. The plane coordinate system movement prediction unit 104 uses the speed of the obstacle from the obstacle detection unit 32 and assumes that the obstacle moves at a constant speed relative to its speed direction, and predicts the movement of the obstacle. By assuming that the obstacle moves at a constant speed, the prediction calculation in the plane coordinate system movement prediction unit 104 can be simplified, and the amount of calculation can be reduced. In addition, the plane coordinate system movement prediction unit 104 predicts the movement of the obstacle in the same cycle as the action cycle of the behavior planning device 102, but when the cycle is sufficiently small, the movement prediction can be performed using only the position of the obstacle in each cycle. In this case, the plane coordinate system obstacle information S3 from the obstacle detection unit 32 is the position of the obstacle.

[0062] The scene determination unit 105 uses the plane coordinate system obstacle movement information S7 from the plane coordinate system movement prediction unit 104, the road information S4 from the road information detection unit 33, and the sensor information S6 from the internal sensor 20 to determine the condition of the obstacle, the road condition, and the driving condition of the vehicle 100, and outputs the condition of the vehicle 100 as scene information S8. Figure 3 The scene determination unit 105 may also use the plane coordinate system obstacle movement information S7 to determine the state of the obstacle, and output the state of the vehicle 100 as the scene information S8. In addition, the scene determination unit 105 may also use the plane coordinate system obstacle movement information S7 and the road information S4 to determine the state of the obstacle and the road state, and output the state of the vehicle 100 as the scene information S8. However, by also using the road information S4 and the sensor information S6, the state can be determined in a wide range.

[0063] In order to determine the driving condition of the vehicle 100, the scene determination unit 105 needs to detect the position of the vehicle 100. Although it can be detected by the internal sensor 20, it can also be detected by the GNSS sensor 27 instead. In this case, the position of the vehicle 100 is output as the vehicle position information S5 from the vehicle position detection unit 34. In addition, the GNSS sensor 27 is one of the external sensors. In addition, the road condition can also be detected by the external sensor. Therefore, the condition of the obstacle, the road condition and the driving condition of the vehicle 100 can all be detected by the external sensor.

[0064] The behavior determination unit 106 determines the behavior of the vehicle 100 based on the scene information S8 from the scene determination unit 105, and outputs a behavior determination result S9. Figure 4 Let me explain in detail.

[0065] The control calculation device 103 includes a motion planning unit 107 and a control calculation unit 108 .

[0066] The action planning unit 107 generates and outputs a target path and a target vehicle speed that the vehicle should travel, using the behavior decision result S9 from the behavior decision unit 106, the reference path S1 and the drivable area S2 from the path detection unit 31, the vehicle position information S5 from the vehicle position detection unit 34, and the sensor information S6 from the internal sensor 20. In addition, the target path and the target vehicle speed are collectively referred to as a target trajectory S10.

[0067] The control calculation unit 108 uses the target trajectory S10 from the motion planning unit 107, the reference path S1 and the drivable area S2 from the path detection unit 31, and the obstacle information from the obstacle detection unit 32 to calculate and output the target steering angle and the target acceleration / deceleration amount S12 so that the vehicle 100 tracks the target trajectory S10.

[0068] The control computing device 103 may not necessarily include the action planning unit 107 when the target trajectory S10 is not generated but the target steering amount S11 and the target acceleration / deceleration amount S12 are directly calculated based on the behavior decision result S9 from the behavior decision unit 106. The control computing device 103 may also calculate the target steering amount S11 and the target acceleration / deceleration amount S12 through model predictive control, etc., based on the behavior decision result S9 from the behavior decision unit 106, the reference path S1 and the drivable area S2 from the path detection unit 31, and the obstacle information from the obstacle detection unit 32. In addition, regarding the actions of the control computing device 103, the following will be used Figure 5 Detailed description.

[0069] Regarding the steering control device 12 and the acceleration / deceleration control device 9, due to the use of Figure 1 This has been explained, so the explanation is omitted here.

[0070] Next, use Figure 3 and Table 1, the scene determination unit 105 is described. Figure 3is a schematic diagram showing an example of a scene in a plane coordinate system in Embodiment 1. Table 1 is an explanatory diagram showing an example of scene information S8 from the scene determination unit 105 in Embodiment 1. The scene information S8 can be expressed as a variable including a numerical value, or can be expressed as a symbol such as scene A and scene B. The following describes a method in which the scene determination unit 105 expresses the scene information S8 as a variable including a numerical value.

[0071] [Table 1]

[0072]

[0073] Set as Figure 3 As shown, within the determination area A1 around the vehicle 100, there are a leading obstacle B1 (leading vehicle), a crossing obstacle B2 (crossing vehicle), a stopped obstacle B3 (stopped vehicle), a signal device C1, a crosswalk C2, and a stop line C3. Here, the determination area A1 is the range in which the scene determination unit 105 determines the conditions of obstacles, road conditions, and the driving conditions of the vehicle 100. That is, the scene determination unit 105 determines these conditions within the determination area A1. The determination area A1 is composed of a plurality of points (in Figure 3 In addition, in order to numerically represent the situation of the vehicle 100, the variables shown in Table 1 are prepared. For example, Figure 3 In the determination area A1, there is an intersecting obstacle B2, so acrobs_inlane = 1. This can be determined based on whether the vector direction of the relative speed of the obstacle to the vehicle 100 is approaching the vehicle 100, according to the plane coordinate system obstacle movement information S7. In addition, by processing the road information S4, that is, the image obtained by the camera 25, it is possible to determine which color signal the signal device C1 represents. In addition, the scene determination unit 105 determines the condition of the obstacle in the determination area A1, the road condition, and the driving condition of the vehicle 100, but is not limited to this. Even if it is an obstacle outside the determination area A1, it can be determined that it is predicted to enter the determination area A1 in the future. In addition, the items for determining the condition are not limited to the items shown in Table 1.

[0074] Next, using Tables 2, 3, and Figure 4 , explaining the behavior determination unit 106. Table 2 is an explanatory diagram showing an example of the behavior determination result S9 from the behavior determination unit 106 in the first embodiment. In addition, Figure 4It is a schematic diagram showing an example of a finite state machine (hereinafter referred to as "FSM (Finite State Machine)") in the behavior determination unit 106 in the first embodiment. In addition, Table 3 is an explanatory diagram showing an example of mode transition in the behavior determination unit 106 in the first embodiment.

[0075] [Table 2]

[0076] project Effectiveness Target Behavior Target path number Reference path information Upper speed limit Lower speed limit Target parking location Target stopping distance

[0077] [Table 3]

[0078]

[0079] Table 2 shows the specific content of the behavior decision result S9 output by the behavior decision unit 106. Validity indicates whether the result determined by the behavior decision unit 106 is valid. This is used by the unillustrated automatic driving device to determine whether to use the behavior planning device 102 to control the vehicle 100 in a scene that the behavior decision unit 106 cannot cope with (for example, scenes outside the automatic driving standard such as the accident scene that is not expected, and scenes where the detection accuracy and reliability of the information acquisition unit 30 are reduced). When the validity is valid, the behavior planning device 102 is used to control the vehicle 100. When the validity is invalid, it means that the scene information S8 from the scene determination unit 105 is inappropriate, so the automatic driving is stopped. The target behavior is the behavior that the vehicle 100 should perform at the current time point or in the future. The target behavior is, for example, driving on the current driving path as it is, or changing the path. The target path number indicates the ID and number assigned to the path that becomes its target when a path change is required. The ID and number are locally assigned based on the section where the vehicle 100 is traveling. Alternatively, it is automatically assigned from the map information S15. The reference path information is information related to the reference path S1. Specifically, the reference path information represents the coordinate values ​​when a point group is used to represent the reference path S1, or represents the parameters when a polynomial, a spline curve, etc. are used to represent the reference path S1. The upper speed limit is a speed based on the legal speed of the vehicle. The lower speed limit is the minimum required speed of the vehicle 100. The target parking position T is the position where the vehicle 100 should stop at the stop line C3, etc. The target parking distance is the distance from the current position of the vehicle 100 to the target parking position T. For these items shown in Table 2, for example, they are expressed in numerical form. In this case, for example, for validity, 1 is assigned when it is valid, and 0 is assigned when it is invalid. The behavior decision unit 106 outputs at least one of the items shown in Table 2 as a behavior decision result S9.

[0080] The control operation device 103 uses the behavior determination result S9 from the behavior determination unit 106 to determine the behavior of the vehicle 100. For example, in the setting of the constraint conditions in the action planning unit 107, the behavior of the vehicle 100 shown in Table 2 is used. When path tracking is set as the behavior of the vehicle 100, the action planning unit 107 generates a target trajectory S10 in a manner such that the vehicle 100 maintains travel within the marking line during travel. In addition, when a path change is set as the target behavior, since it is necessary to cross the marking line, the marking line is removed from the constraint conditions in the action planning unit 107, and the drivable area S2 is expanded to the path of the path change destination to generate the target trajectory S10.

[0081] The behavior decision result S9 from the behavior decision unit 106 is not limited to the example shown in Table 2. It is preferred to set the items of the behavior decision result S9 to match the control operation device 103. For example, in the case where the control operation device 103 requires upper and lower limits of acceleration, steering angle, etc., they may also be included in the items of the behavior decision result S9. In addition, information related to the functions of the control operation device 103, such as the LKS (Lane Keeping System) function as a lane maintenance control, the ACC (Adaptive Cruise Control) function that appropriately controls the distance and relative speed with the vehicle in front, and the TJA (Traffic Jam Assist) function as a function of tracking the vehicle in front, may also be included in the items of the behavior decision result S9.

[0082] Figure 4 Table 3 is an explanatory diagram of a method for the behavior determination unit 106 to output the behavior determination result S9. Specifically, the behavior determination unit 106 uses FSM to determine the behavior. In FSM, a finite number of modes and their transition conditions are first determined. It is preferred to design the modes and their transition conditions according to the scenario of autonomous driving. As an example, assume Figure 4 In addition, the FSM is not limited to Figure 4 The example shown. Figure 4As shown, four modes are set as modes, namely, path tracking (hereinafter referred to as "LF (Lane Following)"), deceleration and stopping (hereinafter referred to as "ST (Stop)"), path change (hereinafter referred to as "LC (Lane Change)"), and emergency stop (hereinafter referred to as "ES (Emergency Stop)"). LF is a mode for traveling on the same path. ST is a mode selected when stopping, such as stopping in front of the stop line C3 or the cross obstacle B2. LC is a mode for changing the path to an adjacent path N. ES is a mode for emergency stopping when there is an obstacle around the vehicle 100.

[0083] like Figure 4 As shown in Table 3, it is designed to enable transition between modes by using scene information S8 from the scene determination unit 105. In Table 3, the current mode represents the current mode of the vehicle 100, and it is assumed that LF is set as the initial mode, that is, starting from LF, when the automatic driving starts. The transfer destination mode is the mode to be transferred next based on the current mode and the transfer condition. The transfer number is a parameter that uses a number to represent the transition from the current mode to the transfer destination mode, and is assigned here (1) to (10). (1) to (10) of Table 3 are respectively Figure 4 The transition condition is a condition in each transition and corresponds to Table 1. The transition formula is a formula that expresses the transition condition using a conditional expression. As in the case of transition number (1), there may be multiple transition formulas. The representative output is an item among the items shown in Table 2, such as an item in which the behavior of the vehicle 100 changes during the transition.

[0084] As an example, the transition number (1) is described. It is assumed that when the vehicle 100 is traveling on the path in the LF mode, there are a signal device C1 and a stop line C3 in front, and the signal device C1 indicates a red signal. In this case, the scene determination unit 105 outputs tgtpоs_inlane=1 (there is a stop line C3 in the determination area A1) and sig_state=2 (the signal device C1 indicates a red signal) to the behavior determination unit 106. The behavior determination unit 106 regards that the transition formula "tgtpоs_inlane==1&&sig_state==2" is satisfied and executes the transition of the transition number (1). That is, the behavior determination unit 106 causes the mode of the vehicle 100 to shift from LF, which is the current mode, to ST. At this time, the behavior determination unit 106 sets the target parking position T to the front of the stop line C3 and calculates the target parking distance, which is output as the behavior determination result S9. In response to the behavior decision result S9 outputted in this way, the motion planning unit 107 in the control computing device 103 generates a target trajectory S10 for stopping the vehicle in front of the stop line C3. Then, the control computing unit 108 calculates the target steering amount S11 and the target acceleration / deceleration amount S12 so that the vehicle 100 follows the target trajectory S10. As a result, the vehicle 100 stops.

[0085] Here, only the transfer number (1) is explained, but the same method is used to determine the transfer destination and determine the behavior of the vehicle 100 for other transfer numbers. In addition, consider a situation where multiple transfer conditions shown in Table 3 are satisfied at the same time. As an example, the current mode is LF and acrobs_inlane=1 (there is an intersecting obstacle B2 in the determination area A1), and req_lc=1 (a route change instruction is transmitted from the navigation device). In this case, it is considered that the transfer destination modes are ST and LC, and it is unclear which one is selected. Therefore, in the case where there are multiple candidates for the transfer destination mode, the transfer destination mode is determined according to a predetermined priority order. For example, in the case where there is an intersecting obstacle B2 in the determination area A1, or in the case where there is an oncoming obstacle B4 in the determination area A1, in order to avoid a collision between the vehicle 100 and the obstacle, a transfer destination mode such as urging the vehicle 100 to stop is preferentially selected.

[0086] As described above, the behavior decision unit 106 considers not only obstacles around the vehicle 100 but also road information S4 such as the signal device C1 to determine the behavior of the vehicle 100. Therefore, the behavior can be determined by considering various conditions around the vehicle 100, which can expand the application range of autonomous driving. In addition, although the method of using FSM in the behavior decision unit 106 is described, it is preferably designed according to the assumed scenario and standard of autonomous driving. Therefore, it is not limited to using Tables 2, 3 and Figure 4The design of FSM is described. In addition, the method of determining the behavior of the vehicle 100 is not limited to FSM. For example, various methods such as a method using a state transition diagram, a method using a neural network or the like to learn in advance, and a method using an optimization technique or the like can be used. In addition, although the case where the scene determination unit 105 outputs a numerical value is described, FSM or the like can be used in the same manner even when the output is expressed using symbols.

[0087] Next, use Figure 5 The operation of the control operation device 103 will be described. Figure 5 This is a schematic diagram showing an example of a target trajectory S10 from the motion planning unit in the first embodiment. Figure 5 This is a specific illustration of the target trajectory S10 output by the action planning unit 107 when avoiding the stopped obstacle B3. Assume that the behavior decision unit 106 determines the behavior of avoiding the stopped obstacle B3 within the drivable area S2 surrounded by the left marking line L1 and the right marking line L2. The action planning unit 107 uses the motion model of the vehicle 100 based on the behavior decision result S9 from the behavior decision unit 106 to more accurately predict the movement of the vehicle 100 and the obstacle. In addition, the action planning unit 107 generates a safe avoidance path within the drivable area S2 as the target trajectory S10 and outputs it to the control operation unit 108. In addition, although in Figure 5 Although not shown, when the vehicle stops because the signal device C1 is red, the action planning unit 107 generates a target trajectory S10 that enables accurate parking at the target parking position T as one of the behavior decision results S9, and outputs it to the control operation unit 108.

[0088] According to the first embodiment described above, the situation of the vehicle 100 is determined based on the obstacle movement information and the road information S4. Therefore, the behavior of the vehicle 100 can be appropriately determined considering not only the obstacle but also the road information S4, and the accuracy of the automatic driving can be improved.

[0089] Implementation method 2.

[0090] Figure 6 This is a block diagram showing an example of the behavior planning device 102 and the control calculation device 103 in the second embodiment. Figure 6 This is a block diagram including the information acquisition unit 30 , the internal sensor 20 , the behavior planning device 102 , the control calculation device 103 , the steering control device 12 , and the acceleration / deceleration control device 9 . Figure 6 and Figure 2The difference between the behavior planning device 102 and the scene planning device 104 is that the behavior planning device 102 further includes a path coordinate transformation unit 109 and a path coordinate system movement prediction unit 110 instead of the plane coordinate system movement prediction unit 104. The components other than the path coordinate transformation unit 109, the path coordinate system movement prediction unit 110 and the scene determination unit 111 are the same as those of the scene planning device 102. Figure 2 The parts shown are the same, so the description is omitted.

[0091] The path coordinate transformation unit 109 transforms the plane coordinate system obstacle information S3 into a path coordinate system based on the reference path S1 and the drivable area S2 from the path detection unit 31 and the plane coordinate system obstacle information S3 from the obstacle detection unit 32, and outputs the plane coordinate system obstacle information S3 as path coordinate system obstacle information S13. When the path coordinate transformation unit 109 transforms the plane coordinate system obstacle information S3, the drivable area S2 is also used, so that the behavior determination unit 106 can determine the behavior of the vehicle 100 taking into account the drivable area S2. However, the drivable area S2 is not necessarily required as an input to the path coordinate transformation unit 109. At least, the path coordinate transformation unit 109 transforms the plane coordinate system obstacle information S3 into a path coordinate system based on the reference path S1 based on the plane coordinate system obstacle information S3 that represents the surrounding obstacles detected by the external sensor provided on the vehicle in a plane coordinate system, and the reference path S1, and outputs the plane coordinate system obstacle information S3 as path coordinate system obstacle information S13. Regarding the path coordinate system, the following will be used. Fig.10 Detailed description.

[0092] The path coordinate system movement prediction unit 110 predicts the movement of the obstacle using the path coordinate system based on the path coordinate system obstacle information S13 from the path coordinate transformation unit 109, and outputs it as the path coordinate system obstacle movement information S14. The path coordinate system movement prediction unit 110 uses the speed of the obstacle from the obstacle detection unit 32, assuming that it is a constant speed linear motion in its speed direction, and predicts the movement of the obstacle. By assuming that it is a constant speed linear motion, the prediction calculation in the path coordinate system movement prediction unit 110 can be simplified, and the amount of calculation can be reduced. In addition, the path coordinate system movement prediction unit 110 predicts the movement of the obstacle in a cycle that is the same as the action cycle of the behavior planning device 102, but when the cycle is sufficiently small, the movement prediction can be performed using only the position of the obstacle in each cycle. In this case, the obstacle information from the obstacle detection unit 32 is the position of the obstacle.

[0093] The scene determination unit 111 uses the path coordinate system obstacle movement information S14 from the path coordinate system movement prediction unit 110, the road information S4 from the road information detection unit 33, and the sensor information S6 from the internal sensor 20 to determine the condition of the obstacle, the road condition, and the driving condition of the vehicle 100, and outputs the condition of the vehicle 100 as the scene information S8. The scene determination unit 111 may also use the path coordinate system obstacle movement information S14 to determine the condition of the obstacle, and output the condition of the vehicle 100 as the scene information S8. In addition, the scene determination unit 111 may also use the path coordinate system obstacle movement information S14 and the road information S4 to determine the condition of the obstacle and the road condition, and output the condition of the vehicle 100 as the scene information S8. However, by also using the road information S4 and the sensor information S6, the condition can be determined in a wide range. The scene determination unit 111 is different from the scene determination unit 111 in that the path coordinate system obstacle movement information S14 is used instead of the plane coordinate system obstacle movement information S7. Figure 2 The scene determination unit 105 shown is different, but has the same function, so the description thereof is omitted.

[0094] Figure 6 The behavior planning device 102 and the control computing device 103 shown in FIG. 1 are installed as the automatic driving system 101. Figure 1 Vehicle 100 is shown.

[0095] Next, use Figure 7 To illustrate the path coordinate system. Figure 7 (a) and (b) are schematic diagrams showing an example of the vehicle 100 in the plane coordinate system and the path coordinate system in the second embodiment. Figure 7 (a) is a schematic diagram in a plane coordinate system. Figure 7 (b) is a schematic diagram of a path coordinate system. The path coordinate system is an LW coordinate system in which the length direction L of the reference path S1 is set as the first axis and the direction W orthogonal to the first axis is set as the second axis. The path coordinate system can usually be transformed from a plane coordinate system.

[0096] Will Figure 7 The representative point Q of the vehicle 100 shown in (a) is transformed to Figure 7 (b) The path coordinate system shown. The representative point Q is, for example, the center of gravity of the vehicle 100, the center of the sensor, etc. Figure 7As shown in (a), when a reference path S1 is provided, an arbitrary point on the reference path S1 is set as the starting point S (preferably the closest point or the rear point of the vehicle 100). From the starting point S along the reference path S1, the length direction of the reference path S1 is set as the L axis, and the axis orthogonal to the reference path S1 is set as the W axis. The intersection point of the perpendicular line from the representative point Q to the reference path S1 and the reference path S1 is set as point P. The length of the reference path S1 from the starting point S along the point P is set as l c , and the length from point P to the representative point Q is set to w c When, the coordinates (x c ,y c ) is transformed into the coordinates of the path coordinate system (l c , w c ). In addition, by Figure 7 The velocity vector V (v x , v y ) is decomposed into the tangent direction component v of the reference path S1 at point P l and its orthogonal component v w , can be used as the velocity in the path coordinate system. Figure 7 As shown in (b), by transforming to the path coordinate system, it is possible to easily determine whether the vehicle 100 is traveling along the path or traveling away from the path. By also transforming the obstacles detected in the plane coordinate system to the path coordinate system, the same determination can be achieved. Thus, by using the reference path S1, the position and speed at any point in the plane coordinate system can be transformed into the position and speed at the point in the path coordinate system. In this way, the path coordinate transformation unit 109 transforms the position and speed of the vehicle 100 and the obstacle from the plane coordinate system to the path coordinate system.

[0097] In addition, when the reference path S1 is set to the center of the driving lane, the path coordinate system can be called the driving lane center coordinate system or the lane coordinate system. In addition, if the markings and the like can also be transformed into path coordinates in the same way, the drivable area S2 can also be expressed in the form of following the path. In this sense, the path coordinate system can be said to be generalized compared to the driving lane center coordinate system and the lane coordinate system.

[0098] Next, use Figure 8 as well as Fig. 9 , an example in which the behavior of vehicle 100 can be more appropriately determined by using a path coordinate system is described. Figure 8 (a) and (b) are schematic diagrams showing an example of the positional relationship between the vehicle 100 and an obstacle when traveling on a curve in the second embodiment. Figure 8 (a) is a schematic diagram in a plane coordinate system. Figure 8(b) is a schematic diagram in the path coordinate system. In addition, Fig. 9 (a) and (b) are schematic diagrams showing an example of the positional relationship between the vehicle 100 and the road information S4 when traveling along a curve in the second embodiment. Fig. 9 (a) is a schematic diagram in a plane coordinate system. Fig. 9 (b) is a schematic diagram in the path coordinate system.

[0099] Figure 8 This represents a scenario where an obstacle is approaching while the vehicle is traveling on a curve. When it is determined that the obstacle intersects the reference path S1 in front of the vehicle 100, it is assumed that the behavior decision unit 106 outputs a behavior decision result S9 in such a way that the vehicle 100 stops. When the movement of the obstacle is predicted in the plane coordinate system, it is assumed that the obstacle moves in a straight line at a constant speed in the plane coordinate system with the detected velocity vector V. In this case, Figure 8 As shown in (a), it is determined that the obstacle crosses at the intersection determination point CR in front of the vehicle 100, and the behavior determination unit 106 outputs the behavior determination result S9 in such a way that the vehicle 100 stops. However, in reality, the obstacle does not directly cross the intersection determination point CR but moves along the shape of the road. When the movement prediction is performed in the plane coordinate system, unnecessary stops often occur, resulting in a poor riding experience or causing discomfort to the occupants. On the other hand, when the movement prediction is performed in the path coordinate system, as shown in FIG. Figure 8 As shown in (b), it is assumed that the obstacle is moving in the direction of the reference path S1 with a speed v l In this case, it is not necessary to determine that the obstacle intersects at the intersection determination point CR in front of the vehicle 100. By performing movement prediction in the path coordinate system, it is possible to suppress frequent stops that occur when the movement prediction is performed in the plane coordinate system.

[0100] Fig. 9 The following is a scenario in which a traffic light C1 is present on a curve and the signal turns red. When the signal C1 turns red, the behavior decision unit 106 outputs the behavior decision result S9 so that the vehicle 100 stops in front of the stop line C3. Furthermore, the behavior decision unit 106 outputs the distance to the stop line C3 as the target stopping distance. Fig. 9 As shown in FIG. 2( a), the distance d1 from the representative point Q of the vehicle 100 to the target parking position T in the plane coordinate system is xy Calculated as dl xy =(dx 2 +dy 2 ) 1 / 2 However, the actual distance is the distance dl along the reference path S1 of the dotted line. r , the distance dl calculated above xyBecome more than the dl r If the value is small, the vehicle may be parked closer to the original target parking position T. On the other hand, if the distance to the target parking position T is measured in the path coordinate system, Fig. 9 As shown in (b), the distance d1 to the target parking position T can be accurately measured. r , the behavior of the vehicle 100 can be determined more accurately. The above describes an example in which the behavior of the vehicle 100 can be determined more appropriately by using the path coordinate system, but the present invention is not limited to Figure 8 as well as Fig. 9 In addition to this, there are also various advantages.

[0101] Fig.10 is a schematic diagram showing an example of a scene in the path coordinate system in Implementation 2. Fig.10 As shown, it is assumed that within the determination area A1 around the vehicle 100, there are a preceding obstacle B1 (preceding vehicle), a crossing obstacle B2 (crossing vehicle), a stopping obstacle B3 (stopping vehicle), an oncoming obstacle B4 (oncoming vehicle), a signal device C1, and a stop line C3. In addition, their positions and speeds are transformed into the path coordinate system by the path coordinate transformation unit 109. In addition, the determination area A1 is set to be composed of the right marking line L2, the left marking line L1, and the predetermined determination distance l l The scene determination unit 111 determines the conditions of obstacles, road conditions, and driving conditions of the vehicle 100 in the determination area A1 around the vehicle 100, and numerically represents the conditions of the vehicle 100 as scene information S8. As variables for numerically representing the scene, the variables shown in Table 1 are prepared. This is the same as the variables used by the scene determination unit 105 in the first embodiment. For example, Fig.10 In the case where there is an intersecting obstacle B2 in the determination area A1, acrobs_inlane = 1. This is done by setting the position of the obstacle to be determined in the path coordinate system to w о And set the speed in the W direction to v wо When we can determine whether w is satisfied о ·v wо < 0. In addition, since the oncoming obstacle B4 exists in the determination area A1, oppos_inlane = 1. This can be determined by whether the speed of the obstacle in the L-axis direction of the determination target in the path coordinate system is negative.

[0102] exist Fig.10 There are right-marked line L2, left-marked line L1, and adjacent path N, but virtually generated parameters may be used without them. Also, the determination area A1 may be set using the travelable area S2 output from the path coordinate conversion unit 109.

[0103] In the second embodiment of the present disclosure, the scene determination unit 111 digitizes the situation of the vehicle 100 in the path coordinate system. And the behavior determination unit 106 determines the behavior of the vehicle 100 based on the digitized scene information S8. Therefore, the inverse transformation from the path coordinate system to the plane coordinate system, which is required when directly calculating the target steering amount S11 based on the movement prediction of the obstacle, is not required, and the calculation load does not need to be increased.

[0104] According to the second embodiment described above, the movement of obstacles is predicted using the obstacle information converted by the path coordinate conversion unit 109, that is, the path coordinate system obstacle information S13. Therefore, the behavior of the vehicle 100 can be determined more appropriately, and the accuracy of automatic driving can be improved.

[0105] Implementation method 3.

[0106] In the third embodiment, a method of performing scene determination in both the plane coordinate system and the path coordinate system will be described. Fig.11 This is a block diagram showing an example of the behavior planning device 102 and the control calculation device 103 in the third embodiment. Fig.11 This is a block diagram including the information acquisition unit 30 , the internal sensor 20 , the behavior planning device 102 , the control calculation device 103 , the steering control device 12 , and the acceleration / deceleration control device 9 . Fig.11 and Figure 2 as well as Figure 6 The difference is that the behavior planning device 102 includes both the plane coordinate system movement prediction unit 104 and the path coordinate system movement prediction unit 110. The components other than the scene determination unit 112 are the same as Figure 2 as well as Figure 6 The parts shown are the same, so the description is omitted.

[0107] The scene determination unit 112 uses the plane coordinate system obstacle movement information S7 from the plane coordinate system movement prediction unit 104, the path coordinate system obstacle movement information S14 from the path coordinate system movement prediction unit 110, the road information S4 from the road information detection unit 33, and the sensor information S6 from the internal sensor 20 to determine the condition of the obstacle, the road condition, and the driving condition of the vehicle 100, and outputs the condition of the vehicle 100 as the scene information S8. The scene determination unit 112 may also use the plane coordinate system obstacle movement information S7 and the path coordinate system obstacle movement information S14 to determine the condition of the obstacle, and output the condition of the vehicle 100 as the scene information S8. In addition, the scene determination unit 112 may also use the plane coordinate system obstacle movement information S7, the path coordinate system obstacle movement information S14, and the road information S4 to determine the condition of the obstacle and the road condition, and output the condition of the vehicle 100 as the scene information S8. However, by also using the road information S4 and the sensor information S6, the condition can be determined in a wide range. In the following, using Fig.12 13, a method in which the scene determination unit 112 uses both the plane coordinate system obstacle movement information S7 and the path coordinate system obstacle movement information S14. Fig.11 The behavior planning device 102 and the control computing device 103 shown in FIG. 1 are installed as the automatic driving system 101. Figure 1 Vehicle 100 is shown.

[0108] Fig.12 (a) and (b) are schematic diagrams showing an example of the positional relationship between the vehicle 100 and an obstacle when traveling at an intersection in the second embodiment. Fig.12 (a) is a schematic diagram in a plane coordinate system. Fig.12 (b) is a schematic diagram in the path coordinate system. In addition, Fig.13 (a) and (b) are schematic diagrams showing an example of the positional relationship between the vehicle 100 and an obstacle when traveling on a T-junction road in the second embodiment. Fig.13 (a) is a schematic diagram in a plane coordinate system. Fig.13 (b) is a schematic diagram in the path coordinate system.

[0109] Fig.12 The following is a scene in which an intersecting obstacle B2 (another vehicle) enters an intersection in the determination area A1. The determination area A1 is composed of a plurality of pre-set points p to s. When an obstacle enters the determination area A1, it is assumed that the behavior determination unit 106 outputs a behavior determination result S9 in such a way that the vehicle 100 stops in front of the intersection. When the movement of the obstacle is predicted in the plane coordinate system, as shown in FIG. Fig.12(As shown in (a)), the intersection points within the obstacle intrusion determination area A1 are correctly determined. On the other hand, in the case where the movement of the obstacle is predicted in the path coordinate system, as Fig.12 (b) shows, points p to s are transformed into the path coordinate system. At this time, they are transformed into the path coordinate system based on the perpendiculars from each point to the reference path S1. However, for point s, there are two perpendiculars, and the intersection points of them become two points, point a and point b. Thus, two transformation point candidates, point sa and point sb, are generated in the path coordinate system. In the plane coordinate system, when the distance from point s to point a is set as w1 and the distance from point s to point b is set as w2, the point corresponding to the smaller one of w1 and w2 becomes the transformation point candidate. Assuming w1 > w2, point sb becomes the transformation point candidate. In this case, the area surrounded by points p, q, r, and sb becomes the determination area A1, and it is determined that the obstacle intrudes into the determination area A1, so it becomes a correct determination result. Assuming w1 < w2, point sa becomes the transformation point candidate. In this case, the area surrounded by points p, q, r, and sa becomes the determination area A1, and it is not determined that the obstacle intrudes into the determination area A1, becoming an incorrect determination result.

[0110] As Fig.12 (a) shows, when multiple points constituting the determination area A1 are respectively transformed into the path coordinate system, in the case where there are multiple candidates for the transformation points corresponding to the points before transformation, an incorrect determination result may occur. At this time, the maximum value of the angle a12 formed by the tangents at two points on the reference path S1 existing within the determination area A1 (hereinafter referred to as "reference angle") becomes larger. In Fig.12 (a), the angle formed by the tangent t1 at point a and the tangent t2 at point b becomes the reference angle, which is approximately 90 degrees. In a normal curve, the reference angle does not become larger, but at an intersection point, the reference angle becomes larger. Therefore, in the case where the reference angle is large, an incorrect determination result may occur. And, in Fig.12In (a), the difference between the distances w1 and w2 is small, so even in such a case, an erroneous determination result may be obtained. Therefore, the scene determination unit 112 distinguishes between the plane coordinate system obstacle movement information S7 and the path coordinate system obstacle movement information S14 according to the scene. When there are two or more transformation points corresponding to the points before transformation to the path coordinate system, the scene determination unit 112 generates scene information S8 based on the plane coordinate system obstacle movement information S7. When there is one transformation point, the scene determination unit 112 generates scene information S8 based on the path coordinate system obstacle movement information S14. Alternatively, when the maximum value of the angle a12 formed by the tangents at two points on the reference path S1 existing in the determination area A1 of the scene determination unit 112 is greater than a predetermined value (for example, 90 degrees), the scene determination unit 112 generates scene information S8 based on the plane coordinate system obstacle movement information S7. When the maximum value of the formed angle is less than a predetermined value, the scene determination unit 112 generates scene information S8 based on the path coordinate system obstacle movement information S14. Alternatively, when there are multiple transformation points, if the difference in length between two perpendicular lines from the point before transformation to the reference path S1 is less than a predetermined value, the scene determination unit 112 generates scene information S8 based on the plane coordinate system obstacle movement information S7. If the difference in length between the two perpendicular lines is greater than a predetermined value, the scene determination unit 112 generates scene information S8 based on the path coordinate system obstacle movement information S14.

[0111] Fig.13 The following is a scene in which a crossing obstacle B2 (bicycle) is approaching from the side of a T-junction and the vehicle 100 is about to turn right. The determination area A1 is composed of a plurality of pre-set points t to w. When it is determined that the obstacle will intersect the reference path S1 in the determination area A1 in the future, it is assumed that the behavior determination unit 106 outputs a behavior determination result S9 in such a way that the vehicle 100 stops. When the movement of the obstacle is predicted in the plane coordinate system, as shown in FIG. Fig.13 As shown in (a), although the obstacle is moving in the direction of the predetermined velocity vector V in the T-junction, it is determined that it does not intersect with the reference path S1, so the correct behavior is determined that the vehicle 100 does not stop but moves parallel to the obstacle. Therefore, it is not necessary to stop in vain. On the other hand, when the movement of the obstacle is predicted in the path coordinate system, as shown in FIG. Fig.13 As shown in (b), points t~w are transformed into the path coordinate system. At the same time, using Fig.13The intersection point c of the perpendicular line from the obstacle to the reference path S1 in (a) with the reference path S1 transforms the position and velocity of the obstacle into the path coordinate system. For points t to w, they are also transformed into the path coordinate system according to the perpendicular lines from each point to the reference path S1, but for point w, there are two perpendicular lines, and the intersection points thereof become two points, namely points d and e. Thus, two transformation point candidates, points wd and we, are generated in the path coordinate system. In the plane coordinate system, when the distance from point w to point d is set as w3 and the distance from point w to e is set as w4, the point corresponding to the smaller one of w3 and w4 becomes the transformation point candidate. Assuming w3 < w4, point wd becomes the transformation point candidate. In this case, the region surrounded by points t, u, v, and wd becomes the determination region A1. Assuming that the intersection point (intersection determination point CR) of the obstacle moving in a straight line at a constant speed with the reference path S1 is not included in the determination region A1, it is determined that the obstacle does not cross the reference path S1 within the determination region A1, so a correct determination result is obtained. Assuming w3 > w4, point we becomes the transformation point candidate. In this case, the region surrounded by points t, u, v, and we becomes the determination region A1. Assuming that the intersection point (intersection determination point CR) of the obstacle moving in a straight line at a constant speed with the reference path S1 is included in the determination region A1, it is determined that the obstacle crosses the reference path S1 within the determination region A1, resulting in an incorrect determination result.

[0112] As Fig.13 (a) shows, when transforming the multiple points constituting the determination region A1 into the path coordinate system respectively, in the case where there are multiple candidates for the transformation points corresponding to the points before transformation, an incorrect determination result may be obtained. At this time, the reference angle becomes larger. In Fig.13 (a), the angle formed by the tangent line t3 at point d and the tangent line t4 at point e becomes the reference angle, but it is approximately 90 degrees. Also in a T-junction, similar to the intersection point, the reference angle becomes larger, so in the case of a large reference angle, an incorrect determination result may be obtained. And, in Fig.13 (a), the difference between the distances w3 and w4 is small, so an incorrect determination result may also be obtained in such a case. Therefore, similar to the case described using Fig.12 , the scene determination unit 112 separately uses the plane coordinate system obstacle movement information S7 and the path coordinate system obstacle movement information S14.

[0113] By separately using the plane coordinate system obstacle movement information S7 and the path coordinate system obstacle movement information S14 according to the scene, the scene determination unit 112 is not limited to Fig.12 and Fig.13In addition, in the conditions for selecting the plane coordinate system obstacle movement information S7 and the path coordinate system obstacle movement information S14, the predetermined value as a comparison object for the maximum value of the angle formed by the tangent at two points and the predetermined value as a comparison object for the difference in length with two perpendicular lines may not be fixed values, but may be set to be variable according to the scene.

[0114] According to the third embodiment described above, by distinguishing and using the plane coordinate system obstacle movement information S7 and the path coordinate system obstacle movement information S14 according to the scene, the behavior of the vehicle 100 can be determined more appropriately, and the accuracy of the automatic driving can be improved.

[0115] Implementation method 4.

[0116] In recent years, each country has been equipped with a high-precision map that distributes high-definition static and dynamic information for autonomous driving of the vehicle 100. Non-patent document 1 describes what kind of information a high-precision map distributes.

[0117] According to non-patent document 1, a high-precision map is a map formed by superimposing dynamic data classified by information update frequency on a basic map (static information). Dynamic data is classified into quasi-static information, quasi-dynamic information, and dynamic information. For quasi-static information, the update frequency is within 1 day, including, for example, traffic control information and road construction information. For quasi-dynamic information, the update frequency is within 1 hour, including, for example, accident information and congestion information. For dynamic information, the update frequency is within 1 second, including, for example, signal information and pedestrian information.

[0118] High-precision maps generally use geographic coordinate systems, similar to the GNSS sensor 27. Figure 1 It is generally composed of wide-area data of several hundred km units. Therefore, by using a high-precision map, it is possible to obtain information of a wider area in advance, and the behavior of the vehicle 100 can be determined by considering a wide range of conditions. In addition, the high-precision map includes a basic map and dynamic data, so obstacle information can also be obtained. Therefore, by combining with the GNSS sensor 27, the information acquisition unit 30 can be made into a simple structure. In addition, the accuracy of the behavior decision result S9 output by the behavior planning device 102, and the accuracy of the target steering amount S11 and the target acceleration and deceleration amount S12 output by the control operation device 103 are improved.

[0119] Fig.14 This is a block diagram showing an example of the behavior planning device 102 and the control calculation device 103 in the fourth embodiment. Fig.14This is a block diagram including the information acquisition unit 30 , the internal sensor 20 , the behavior planning device 102 , the control calculation device 103 , the steering control device 12 , and the acceleration / deceleration control device 9 . Fig.14 and Figure 2 The difference between the two is that a high-precision map acquisition unit 35 is provided instead of the path detection unit 31, the obstacle detection unit 32, and the road information detection unit 33. The components other than the high-precision map acquisition unit 35 are the same as Figure 2 The parts shown are the same, so the description is omitted.

[0120] The high-precision map acquisition unit 35 acquires a high-precision map, and outputs a reference path S1, a drivable area S2, map information S15, plane coordinate system obstacle information S3, and road information S4. This information is information in a plane coordinate system. Therefore, the high-precision map acquisition unit 35, like the vehicle position detection unit 34, has a function of transforming information represented by a geographic coordinate system into a plane coordinate system. In addition, the plane coordinate system obstacle information S3 may not necessarily be output from the high-precision map acquisition unit 35, but may also be output from the obstacle detection unit 32. In addition, the information acquired by the high-precision map acquisition unit 35 and the vehicle position detection unit 34 is information represented by a geographic coordinate system, but is not limited to the geographic coordinate system. In addition, the high-precision map acquisition unit 35 can also be applied to Figure 2 , Figure 6 as well as Fig.13 The behavior planning device 102 and the control operation device 103 are shown.

[0121] Fig.14 The behavior planning device 102 and the control computing device 103 shown in FIG. 1 are installed as the automatic driving system 101. Figure 1 Vehicle 100 is shown.

[0122] According to the fourth embodiment described above, by using a high-precision map, it is possible to improve the accuracy of the behavior decision result S9 output by the behavior planning device 102 and the accuracy of the target steering amount S11 and the target acceleration / deceleration amount S12 output by the control calculation device 103. Thus, the accuracy of the automatic driving can be improved.

[0123] Furthermore, the behavior planning device 102 and the control computing device 103 in the first to fourth embodiments are described as being applied to the automatic driving of the vehicle 100, but the application is not limited to the automatic driving, and can be applied to various mobile bodies. For example, it can be applied to mobile bodies that require safe operation, such as mobile robots that conduct inspections in buildings, production line inspection robots, and personal mobile devices.

[0124] Explanation of symbols

[0125] 1: Steering wheel; 2: Steering shaft; 3: Electric motor; 4: Rack and pinion mechanism; 5: Tie rod; 6: Front steering knuckle; 7: Vehicle drive unit; 8: Transmission shaft; 9: Acceleration and deceleration control unit; 10: Brake control unit; 11: Brake; 12: Steering control unit; 13: Pinion shaft; 14: Rack shaft; 15: Front wheel; 16: Rear wheel; 20: Internal sensor; 21: Vehicle speed sensor; 22: IMU sensor; 23: Steering angle sensor; 24: Steering torque sensor; 25: Camera; 26: Radar; 27: GNSS sensor; 28: Navigation device; 29: LiDAR; 30: Information acquisition unit; 31: Path detection unit; 32: Obstacle detection unit; 33: Road information detection unit; 34: Vehicle position detection unit; 35: High-precision map acquisition unit; 100: Vehicle; 101: Autonomous driving system; 102: Behavior planning device; 103: Control computing device; 104: Plane coordinate system movement prediction unit; 105, 111, 112: Scene Determination unit; 106: behavior determination unit; 107: action planning unit; 108: control operation unit; 109: path coordinate transformation unit; 110: path coordinate system movement prediction unit; S1: reference path; S2: drivable area; S3: plane coordinate system obstacle information; S4: road information; S5: vehicle position information; S6: sensor information; S7: plane coordinate system obstacle movement information; S8: scene information; S9: behavior determination result; S10: target track; S11: target steering amount ; S12: target acceleration and deceleration; S13: path coordinate system obstacle information; S14: path coordinate system obstacle movement information; S15: map information; A1: judgment area; B1: preceding obstacle; B2: crossing obstacle; B3: stop obstacle; B4: oncoming obstacle; C1: signal equipment; C2: pedestrian crossing; C3: stop line; L1: left marking line; L2: right marking line; N: adjacent path; Q: representative point; S: starting point; T: target parking position; CR: intersection judgment point.

Claims

1. A behavior planning device, comprising: a plane coordinate system movement prediction unit that predicts the movement of the obstacle based on plane coordinate system obstacle information that represents the obstacle in a plane coordinate system, and outputs the predicted movement as plane coordinate system obstacle movement information; A scene determination unit, which determines the state of the obstacle according to the obstacle movement information of the plane coordinate system, and outputs the state of the moving body as scene information expressed by numerical values ​​or symbols; a behavior determination unit that determines the behavior of the moving body based on the scene information expressed as a numerical value or a symbol, and outputs the behavior determination result to a control operation device that calculates a target value for controlling the moving body according to the behavior determination result; a path coordinate transformation unit, which transforms the plane coordinate system obstacle information into a path coordinate system based on the reference path according to the plane coordinate system obstacle information and a reference path as a driving reference of the mobile body, and outputs the transformed information as path coordinate system obstacle information; as well as a path coordinate system movement prediction unit, which predicts the movement of the obstacle according to the path coordinate system obstacle information and outputs the predicted movement of the obstacle as the path coordinate system obstacle movement information; The scene determination unit determines the state of the obstacle based on the plane coordinate system obstacle movement information and the path coordinate system obstacle movement information, and outputs the state of the moving object as scene information.

2. The behavior planning device according to claim 1, wherein: The scene determination unit determines the state of the obstacle and the road state based on the plane coordinate system obstacle movement information, the path coordinate system obstacle movement information, and the road information, and outputs the state of the moving body as the scene information.

3. The behavior planning device according to claim 1 or 2, wherein: The path coordinate conversion unit converts a plurality of points that are preset to form the determination area of ​​the scene determination unit into a path coordinate system and outputs the converted points. When there are two or more transformation points corresponding to the point before transformation to the path coordinate system, the scene determination unit generates the scene information based on the obstacle movement information in the plane coordinate system, When the number of the conversion points corresponding to the point before conversion to the path coordinate system is one, the scene determination unit generates the scene information based on the path coordinate system obstacle movement information.

4. The behavior planning device according to claim 1 or 2, wherein: When the maximum value of the angle formed by tangent lines at two points on the reference path existing in the determination area of ​​the scene determination unit is greater than a predetermined value, the scene determination unit generates the scene information based on the obstacle movement information in the plane coordinate system, When the maximum value of the formed angle is equal to or smaller than the predetermined value, the scene determination unit generates the scene information based on the path coordinate system obstacle movement information.

5. The behavior planning device according to claim 1, wherein: The plane coordinate system movement prediction unit predicts the movement of the obstacle based on the position of the obstacle or both the position and speed of the obstacle as the plane coordinate system obstacle information.

6. The behavior planning device according to claim 1 or 2, wherein: The plane coordinate system movement prediction unit predicts the movement of the obstacle based on the position of the obstacle or both the position and speed of the obstacle as the plane coordinate system obstacle information. The path-coordinate-system movement prediction unit predicts the movement of the obstacle based on the position of the obstacle or both the position and speed of the obstacle as the path-coordinate-system obstacle information.

7. The behavior planning device according to claim 1, wherein: The plane coordinate system movement prediction unit performs movement prediction on the assumption that the obstacle moves linearly at a constant speed.

8. The behavior planning device according to claim 1 or 2, wherein: The plane coordinate system movement prediction unit performs movement prediction under the assumption that the obstacle moves in a straight line at a constant speed. The path coordinate system movement prediction unit performs movement prediction on the assumption that the obstacle moves linearly at a constant speed.

9. The behavior planning device according to claim 1 or 2, wherein: The behavior determination unit determines the behavior of the moving object using a finite state machine.

10. The behavior planning device according to claim 1 or 2, wherein: The behavior decision result from the behavior decision unit is at least one of validity indicating whether the behavior decision result is valid, a target behavior of the mobile body, a target path number pre-assigned to a target path of the mobile body, reference path information related to a reference path serving as a driving basis for the mobile body, an upper limit speed based on a legal speed of the mobile body, a lower limit speed which is a minimum required speed of the mobile body, a target parking position which is a position where the mobile body should park, and a target parking distance which is a distance from a current position of the mobile body to the target parking position.

11. A behavior planning device, comprising: a path coordinate transformation unit that transforms the plane coordinate system obstacle information into a path coordinate system based on the reference path, based on the plane coordinate system obstacle information representing the obstacle in the plane coordinate system and the reference path as the driving reference of the moving body, and outputs the plane coordinate system obstacle information as the path coordinate system obstacle information; a path coordinate system movement prediction unit, which predicts the movement of the obstacle according to the path coordinate system obstacle information and outputs the prediction as the path coordinate system obstacle movement information; a scene determination unit that determines the state of the obstacle based on the path coordinate system obstacle movement information and the plane coordinate system obstacle movement information obtained by predicting the movement of the obstacle based on the plane coordinate system obstacle information, and outputs the state of the moving body as scene information represented by a numerical value; as well as The behavior determination unit determines the behavior of the moving body based on the scene information expressed as numerical values, and outputs the behavior determination result to a control calculation device that calculates a target value for controlling the moving body according to the behavior determination result.

12. The behavior planning device according to claim 11, wherein: The scene determination unit determines the state of the obstacle and the road state based on the path coordinate system obstacle movement information, the plane coordinate system obstacle movement information, and the road information, and outputs the state of the moving body as the scene information.

13. The behavior planning device according to claim 11 or 12, wherein: The path-coordinate-system movement prediction unit predicts the movement of the obstacle based on the position of the obstacle or both the position and speed of the obstacle as the path-coordinate-system obstacle information.

14. The behavior planning device according to claim 11 or 12, wherein: The path coordinate system movement prediction unit performs movement prediction on the assumption that the obstacle moves linearly at a constant speed.

15. The behavior planning device according to claim 11 or 12, wherein: The behavior determination unit determines the behavior of the moving object using a finite state machine.

16. The behavior planning device according to claim 11 or 12, wherein: The behavior decision result from the behavior decision unit is at least one of validity indicating whether the behavior decision result is valid, a target behavior of the mobile body, a target path number pre-assigned to a target path of the mobile body, reference path information related to a reference path serving as a driving basis for the mobile body, an upper limit speed based on a legal speed of the mobile body, a lower limit speed which is a minimum required speed of the mobile body, a target parking position which is a position where the mobile body should park, and a target parking distance which is a distance from a current position of the mobile body to the target parking position.

17. A control operation device that performs operation of the target value according to the behavior determination result output from the behavior planning device according to any one of claims 1 to 16.

Images