Driving Support Device and Driving Support Method

Abstract

Provided are a driving assistance device and a driving assistance method that enable the timely determination of when to stop driving assistance / automated driving due to a geographical feature error in a map. The driving assistance device is characterized by comprising: a determination point setting unit that sets a determination point having associated therein an assumed self-position of a vehicle assumed on a map, the position on the map of a measured geographical feature measurable from the assumed self-position, and an error predicted as the difference between the actual position and the position of the measured geographical feature on the map; a target geographical feature measurement unit that determines the measurement position of the measured geographical feature on the basis of external environment information acquired by an external environment sensor mounted to the vehicle; a self-position estimation unit that estimates the self-position of the vehicle on the map on the basis of the external environment information; and a driving assistance state setting unit that sets the control mode of a driving assistance system for the vehicle on the basis of the determination point, the measurement position of the measured geographical feature measured by the target geographical feature measurement unit, and the self-position of the vehicle estimated by the self-position estimation unit.

Description

Technical Field

[0001] The present invention relates to a driving support device and a driving support method for a vehicle.

Background Art

[0002] In recent years, the development of driving support devices and driving support methods for vehicles using maps has been actively promoted, and Patent Document 1 is known as an example of this.

[0003] In Patent Document 1, when determining whether autonomous driving control is possible according to the estimation accuracy of self-position estimation required at a point ahead on the planned travel route, it is suppressed that it is unnecessarily determined that autonomous driving control is impossible before the point where the estimation accuracy is required. For this purpose, "a driving support method is provided in which the planned travel route of the host vehicle is set, the self-position that is the current position of the host vehicle is estimated, and the driving behavior of the host vehicle is controlled so as to support travel along the planned travel route based on the planned travel route and the self-position. In this driving support method, the estimation error of the self-position is estimated, and the allowable error of the self-position corresponding to the driving behavior of the host vehicle planned at a point on the planned travel route ahead of the self-position is set, and the allowable error of the self-position is larger the farther the point is from the self-position if the driving behavior of the host vehicle is the same. Whether or not to control the driving behavior of the host vehicle is determined according to whether the estimation error of the self-position is within the allowable error." It is configured as follows.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] According to Patent Document 1, when determining whether autonomous driving control is possible according to the allowable error of the self-position required at a point ahead of the self-position on the planned travel route, it is possible to suppress the unnecessary determination that autonomous driving control is impossible before that point.

[0006] However, according to Patent Document 1, when the accuracy of the map is low, it may become impossible to continue driving support / autonomous driving at inappropriate timings. To show this in a more straightforward example, based on a pre-determined planned travel route and self-position, especially when turning right while driving on the left side, regarding the position of the target object in the right front direction, it is assumed that it will be found during a right turn that the actually measured position of the target object is closer to the self-position than the assumed position on the planned travel route.

[0007] In such a state, there is a risk that it may become impossible to continue autonomous driving, such as when the vehicle needs to reverse, and reversing within an intersection may cause safety problems. Or, although it may be possible to continue driving by making an emergency correction to the planned travel route, there is a risk of an uncomfortable situation such as a deterioration in the riding comfort of the passengers.

[0008] From the above, an object of the present invention is to provide a driving support device and a driving support method that can determine at an appropriate timing the stop of driving support / autonomous driving due to the object error of the map.

Means for Solving the Problem

[0009] Therefore, in the present invention, "a determination point setting unit that sets a determination point associating a hypothetical self-position of the host vehicle assumed on a map, a position on the map of a measurement target feature that can be measured from the hypothetical self-position, and an error predicted as a difference between the position on the map of the measurement target feature and the actual position; a target feature measurement unit that obtains a measurement position of the measurement target feature based on external information acquired by an external sensor mounted on the host vehicle; a self-position estimation unit that estimates the self-position of the host vehicle on the map based on the external information; and a driving support state setting unit that sets a control mode of the driving support system of the host vehicle based on the determination point, the measurement position of the measurement target feature measured by the target feature measurement unit, and the self-position of the host vehicle estimated by the self-position estimation unit."

[0010] Further, in the present invention, "a driving support method in which a computer device generates a driving track of a vehicle on a map and supports the driving of the vehicle along the driving track, and on the driving track generated using the position of a feature on the map, a position on the limit driving track that enables stable driving when driving according to a given vehicle motion allowance is set as a determination point in consideration of the feature error, and when the vehicle is actually running, it is determined whether driving support is possible according to the measured feature position when the measured vehicle position reaches the determination point."

Effects of the Invention

[0011] It is possible to provide a driving support device and a driving support method capable of determining, at an appropriate timing, the stop of driving support / automatic driving due to the feature error of the map.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Embodiments for Carrying Out the Invention

[0013] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

Embodiment

[0014] FIG. 1 is a diagram showing a configuration example of a driving support device according to Embodiment 1 of the present invention. The driving support device 10 mounted on the vehicle 1 is roughly classified into a trajectory creation processing unit 11 that determines a planned travel route (hereinafter referred to as a travel trajectory) in advance, and a trajectory travel confirmation processing unit 12 that confirms that the vehicle can continue to travel along the travel trajectory during actual travel. Note that the driving support device 10 is configured using a computer device.

[0015] Among these, the trajectory creation processing unit 11 stores the vehicle motion allowance D1 and the map information D2 in a storage unit such as a ROM in advance, and the trajectory generation unit 16 creates a travel trajectory on the map in advance using these pieces of information. In the present invention, in the determination point setting unit 15, determination points, which will be described in detail later, are set on this travel trajectory.

[0016] FIG. 2 is a diagram illustrating the relationship between the travel trajectory and the determination point. According to the travel trajectory PL created by the trajectory generation unit 16 in the example of the figure, a vehicle traveling on the left side as in Japan is planned to turn right at an intersection.

[0017] The method of creating the travel trajectory PL in the trajectory generation unit 16 is a well-known method, and the description thereof will be omitted. Briefly speaking, as shown in FIG. 2, using the feature position D2a included in the map information D2, the travel trajectory PL is created so that the vehicle does not contact the feature position D2a. The feature position D2a in this example is the position of the left end of the road, the white line or the curb at the corner of the intersection, and other things such as the white line of the lane are also included in this.

[0018] Note that, as the map, for example, an SD map that is widely maintained even on general roads is used. The SD map has a larger feature error D2b compared to the high-precision map. When the trajectory generation unit 16 creates the travel trajectory PL, the travel trajectory PL is considered in consideration of the feature error D2b with respect to the feature position D2a in the map information. Also, as the map, a map (self-generated map) generated from the measurement results of the sensors mounted on the vehicle 1 may be used.

[0019] In the present invention, for the travel trajectory PL on the map created as described above, the determination point Dp shown in FIG. 2 is set in the determination point setting unit 15. The determination point Dp in this example is a point on the travel trajectory PL that defines the limit point that enables traveling according to a preset vehicle movement allowance D1, for example, according to the turning radius.

[0020] Incidentally, if it is before the determination point Dp, even if there is a ground feature error D2b, the vehicle can travel by making a gentle turn. However, if it is found that there is a ground feature error D2b at a position past the determination point Dp, a sharp turn becomes necessary. In some cases, not only a sharp turn is required, but there is also a risk of contact with a ground feature (specifically referred to as a measurement target ground feature) in the right front. The critical point that leads to such an event is defined as the determination point Dp. The determination point Dp is defined as information including the position on the travel trajectory PL and the position of the measurement target ground feature.

[0021] In this way, in the trajectory creation processing unit 11, the determination point Dp composed of the position on the map and the position of the measurement target ground feature is set from the vehicle movement allowance D1 and the ground feature position D2a and error D2b of the map. It can be said that this determination point Dp defines the limit point for continuing the travel on the travel trajectory PL in order to give the passengers a sense of security or protect safety.

[0022] In the above description, the concept of the processing of the determination point setting unit 15 and the trajectory generation unit 16 has been explained. However, these processes are handled virtually on the map. The determination point setting unit 15 can be said to "set a determination point in association with the assumed self-position of the own vehicle assumed on the map, the position of the measurement target ground feature on the map that can be measured from the assumed self-position, and the error predicted as the difference between the position of the measurement target ground feature on the map and the actual position."

[0023] Furthermore, if the processing of the determination point setting unit 15 is described in detail, this is "when the trajectory generation unit cannot generate a travel trajectory that satisfies the vehicle movement allowance of the own vehicle, the assumed self-position as the starting point of the travel trajectory and the position and error of the measurement target ground feature on the map that can be measured from the assumed self-position are associated and set as the determination point."

[0024] Also, the trajectory generation unit 16 is "to generate a travel trajectory of the own vehicle starting from the assumed self-position based on the assumed self-position on the map, the position and error of the measurement target ground feature on the map."

[0025] On the other hand, in the trajectory travel confirmation processing unit 12 of FIG. 1, among various sensors 17 provided in the vehicle 1, using the output (external information) of the external sensor, in the target object measurement unit 18, among a plurality of objects along with the vehicle travel, for example, when turning right during left-side travel, the object to be monitored is recognized as the measurement target object, and the distance, position, etc. are determined. Also, among various sensors 17 provided in the vehicle 1, for example, using the external information, the self-position estimation unit 19 estimates the travel position (self-position estimation result) of the vehicle 1.

[0026] In addition, in the driving support state setting unit 13 within the trajectory travel confirmation processing unit 12, the control mode of driving support / automatic driving is set from the travel trajectory PL created in advance by the trajectory creation processing unit 11 and the current situation (external recognition result, self-position estimation result, determination point) confirmed in the trajectory travel confirmation processing unit 12. The control mode is, for example, whether automatic driving can continue, and when the position of the determination point Dp is exceeded and the measurement target object is not recognized, it is determined that the continuation of automatic driving is not possible.

[0027] As described above, the concept of the processing of the trajectory travel confirmation processing unit 12 has been explained. The trajectory travel confirmation processing unit 12 can be said to be one that "sets the control mode of the driving support system of the own vehicle based on the determination point, the measurement position of the measurement target object measured by the target object measurement unit, and the self-position of the own vehicle estimated by the self-position estimation unit."

[0028] FIG. 3 is a flowchart showing the processing content of the determination point setting unit 15. In this flowchart, three sets of loop processes are set. One of them is a loop process that repeatedly executes processing for objects, and in particular, for objects located on the left side of the travel trajectory, the processing from processing step S1 to processing step S11, which is sequentially called for these, corresponds to this.

[0029] This loop is a large loop, and within this, there is a loop process that repeatedly executes processing for the direction of the map error as a middle loop, and the processing from processing step S2 to processing step S9 is sequentially executed by sequentially calling the direction of the map error.

[0030] For this inner loop, there is a loop process that further repeatedly executes the process for sampling points as a small loop inside it, sequentially calls the sampling points, and sequentially executes the processes within processing steps S4 to S8.

[0031] Thus, in the flow of FIG. 3, a plurality of ground objects D2a on the driving trajectory are sequentially called in processing step S1. For example, assume that the ground object in the upper right of FIG. 4 is being processed. This example in the upper right is a right-turning scene at an intersection and is the most notable scene for autonomous driving judgment. Hereinafter, this ground object is referred to as the measurement target ground object. Next, in processing step S2, the map error D2b included in the map position of the selected measurement target ground object D2a is processed. Here, for each ground object, it is sequentially processed while referring to the map error in the area.

[0032] In processing step S3, based on the direction of the ground object and the map error under the extracted conditions, the ground object position and the target position are calculated. For example, FIG. 4 is a diagram showing the processing content of FIG. 3. D2a shown by the dotted line is the position of the measurement target ground object on the map D2. If the magnitude and direction of the map error D2b in the vicinity thereof are as shown in the figure, the ground object position is changed like the solid line D2a'. In this case, there is a concern about interference with the modified ground object position D2a' in the driving trajectory PL initially planned by the trajectory generation unit 16. Therefore, in processing step S3, the trajectory generation unit 16 further determines the target position Po after the right turn in the driving trajectory PL initially planned as the modified target position Pn considering the magnitude and direction of the map error D2b.

[0033] In processing step S4, the process of the sampling point SP set on the driving trajectory PL of FIG. 4 is performed. A plurality of sampling points SP are set at intervals, and the determination process is sequentially performed from the front in the figure. In this example, first, the process of sampling point SP1 is performed, and in processing step S5, the trajectory from the sampling point SP1 to the target position Po is calculated. However, if the target position Po has been changed to the modified target position Pn as a result of considering the magnitude and direction of the map error D2b, the trajectory from the sampling point SP1 to the modified target position Pn is calculated.

[0034] In processing step S6, it is determined whether the vehicle can calculate the trajectory to the target position Po (or the corrected target position Pn) based on the turning radius set as the vehicle movement allowance D1. When it can be calculated, the process returns from processing step S8 to processing step S4, and the same processing and determination are repeated for the next sampling point SP2.

[0035] As a result, it is assumed that the trajectory calculation is successful up to the sampling point SP5, but the trajectory calculation cannot be performed at the sampling point SP6. In this case, in processing step S7, the sampling point SP6 at which the trajectory calculation could not be performed is recognized as the determination point Dp, and the determination point Dp at this time is recorded as information including the position on the travel trajectory PL and the position of the measurement target feature D2a. Note that the sampling point SP5 before the sampling point SP6 at which the trajectory calculation could not be performed may be recognized as the determination point Dp.

[0036] The above repeated processing in FIG. 3 is sequentially determined for a plurality of features D2a on the travel trajectory PL, and is repeatedly executed until the processing for all the target features is completed. In processing step S10, the determination point closest to the vehicle on the travel trajectory PL is selected.

[0037] According to the above processing, at the final sampling point SP5 before the determination point Dp (sampling point SP6), it is determined that the vehicle can calculate the trajectory to the target position Po (or the corrected target position Pn), and the trajectory cannot be calculated at the determination point Dp. Regarding the handling of the travel trajectory PL at this time, the following two methods are conceivable, and the present invention may adopt either of them.

[0038] One method is to keep the travel trajectory PL with the information of the determination point Dp, and the other method is to add the information of the determination point Dp as a new travel trajectory PL created at the corrected target position Pn.

[0039] Returning to FIG. 1, the travel trajectory PL provided by the trajectory creation processing unit 11 includes information on the determination point Dp considering the above map error D2b. Moreover, in the trajectory travel confirmation processing unit 12, when automatically driving on the travel trajectory PL including information on the determination point Dp considering the map error D2b, the current situation is judged from the external recognition result and the self-position estimation result detected by the sensor 17, and it is confirmed at a stage before reaching the determination point Dp that the driving operation can be performed safely without giving the occupant anxiety, and the control mode of driving support / automatic driving is set. The control mode is, for example, whether automatic driving can be continued, and when the position beyond the determination point Dp is exceeded and the measurement target feature is not recognized, it is judged that the continuation of automatic driving is impossible.

[0040] FIG. 5 is a diagram for explaining a measure for quickly transferring from automatic driving to the driver when it is judged that the continuation of automatic driving is impossible. Multiple settings are possible on the safe side for the countermeasure in such a situation.

[0041] In FIG. 5, Dp is a determination point based on trajectory generation, and passing this point may impede stable and safe operation along the travel trajectory PL. Therefore, when the determination point Dp approaches, it is necessary to notify the driver in advance and prepare the driver for manual driving.

[0042] As this countermeasure, the determination point setting unit 15 sets, as a limit point, the determination point Dp, and also sets a determination point for notification or system preparation on the travel trajectory PL at a position before the determination point Dp as the limit point.

[0043] Among these, when the determination point at which the driver's manual driving posture is completed and thus the transfer from automated driving to the driver is made is defined as the transfer point Pc from automated driving to the driver, it is conceivable to set Pc1 and Pc2 in front of the determination point Dp. The transfer point Pc1 is the position where the steering angle of the steering wheel becomes 0 degrees, and since this position may be after entering the intersection, the transfer point Pc2 is set as the position in front of the change in the road structure in front of the intersection entrance.

[0044] In addition, when setting the transfer point Pc in the determination point setting unit 15, it is preferable to set the position of the transfer point Pc in front of the determination point Dp based on at least one of the steering angle of the host vehicle at the assumed own position, the switching of the road area in the map, and the reaction time of the driver.

[0045] Regarding the determination point (hereinafter referred to as the notification point) for guiding and notifying the driver, since the transfer point Pc assumes the position when the driver becomes operable, as the timing of the notification point for guiding the driver of the switch from the driving support device, it is preferable to switch to driver operation at the transfer point Pc as T seconds before the determination point (transfer point Pc), or to give a notice of the possibility of switching T + α seconds before.

[0046] According to the correspondence in FIG. 5, by modifying to include the pre-point so that the determination point Dp based on the trajectory generation becomes optimal for the driver, there is an effect that the driving can be smoothly transferred to the driver.

[0047] FIG. 6 is a diagram for explaining the setting of a plurality of types of determination points Dp. In the description up to FIG. 5, the determination point Dp has been described with one place in mind, but in FIG. 6, a plurality of levels of determination points Dp1 and Dp2 are calculated. In this figure, the determination point Dp2 on the front side is determined from the viewpoint of the sense of security given to the passenger that there is a possibility of wobbling if this point is exceeded, and the determination point Dp1 during turning is determined from the viewpoint that if this point is exceeded, there is a possibility that backward movement is required, and thus there is a possibility that the minimum turning radius cannot be turned around.

[0048] In this case, in the section between the determination points Dp1 and Dp2, it is preferable to adopt a control mode in which the speed is reduced in preparation for a turn with a small turning radius R, and in the section after the determination point Dp1, it is preferable to adopt a control mode in which the driving is transferred to the driver.

[0049] Therefore, the information of the determination point Dp is preferably defined as information including the position on the travel trajectory PL, the position of the measurement target feature, and further the control mode in the section between the determination points Dp1 and Dp2.

[0050] When calculating a plurality of levels of determination points Dp1 and Dp2, a plurality of combinations of the vehicle motion allowance and the speed limit may be prepared. For example, by changing at least one of the determination condition for successful trajectory calculation and the vehicle motion allowance D1 so that the positions are different between the determination point Dp when entering the intersection relatively fast and the determination point Dp when entering the intersection relatively slowly, a plurality of levels of determination points can be calculated, and the control mode can be set according to the level of the determination point. Thereby, there is an effect that the control mode can be set appropriately.

[0051] Note that the determination point setting unit 15 may allow only forward movement as the vehicle motion allowance D1 and not allow backward movement. In this case, it is preferable to set the determination point Dp for determining that the driving support cannot be continued and the driving is transferred to the driver with the possibility of backward movement.

[0052] In addition, when using the turning radius of the host vehicle as the vehicle motion allowance D1, it can be treated as a variable value such that the upper limit speed of the driving support system is set to be smaller as the turning radius becomes smaller.

Embodiment

[0053] In the first embodiment, the travel path PL is created with respect to the ground features, whereas in the second embodiment, the determination is made taking into account the road conditions in addition to the ground features.

[0054] FIG. 7 is a flowchart showing the processing content of the determination point setting unit 15 that takes into account the road conditions. In the repetitive process of FIG. 3, the repetitive process is performed with respect to the direction of the ground features and the map error, whereas in the second embodiment, the repetitive process is performed with respect to the road conditions.

[0055] In this case, it is a repetition inside the processing step S20 and the processing step S21, but since this internal processing is the same as that of FIG. 3 in the first embodiment, the description thereof is omitted. Briefly, in the processing step S20, a temporary obstacle is set as the road condition, and the travel path PL and the determination point Dp are set with respect to this temporary obstacle. That is, the determination point setting unit 15 sets the determination point based on the position of the assumed obstacle on the map instead of the position and error of the measurement target ground feature on the map.

[0056] As the obstacle to be set temporarily, for example, it is assumed that vehicles 1X are parked at regular intervals along the road boundary, and in particular, when assuming a situation where a bus is parked at a bus stop, etc., it is advisable to set the travel path PL and the determination point Dp in such a situation.

[0057] FIG. 8 is a diagram showing the processing content of FIG. 7, and the travel path PL and the determination point Dp are set when there is a temporary vehicle 1X at each sampling point SP. When the trajectory cannot be calculated, it is set as the determination point Dp.

[0058] At this time, when the driving support continuation determination unit 13 exceeds the position of the determination point Dp and it is unknown whether there is no obstacle at the temporary obstacle position, it is determined that the continuation of the automatic driving is impossible. This is realized, for example, when the driving support state setting unit cannot recognize that there is no assumed obstacle corresponding to the assumed self-position after the self-position and the assumed self-position match, and it is determined that the automatic driving by the driving support system of the own vehicle cannot be continued.

Embodiment

[0059] In the first embodiment, the ground object error D2b is handled as information included in the map (SD map or self-generated map), but this may also be handled as an error based on actual measurement.

[0060] FIG. 9 is a diagram showing handling the error based on actual measurement as the ground object error. In this figure, for example, it is assumed that the ground object position D2a estimated by the trajectory creation unit 11 in FIG. 1 based on the position information of the map is different from the ground object position D2a' measured by the target ground object measurement unit 18 at the point where the vehicle has already passed. And it is assumed that the magnitude and direction of the ground object error at this time are known as the measurement result.

[0061] In the third embodiment of the present invention, if so, the map error as the measurement result in the past is handled as the map error at a future point. Specifically, it is used for setting the driving trajectory PL and the determination point Dp at an intersection where a right turn is predicted from now on. In this case, regarding which of the map error obtained from the map and the map error obtained from the measurement result should be used, or whether both should be used, it can be appropriately determined and applied.

[0062] This process is to set the magnitude of the error of other measurement target ground objects existing around the own vehicle by comparing the measurement position of the measurement target ground object measured by the target ground object measurement unit 18 with the position of the measurement target ground object on the map in the determination point setting unit 15.

Embodiment

[0063] In the above embodiments, when configuring the driving assistance device 10 using a computer device, the computational load is not considered. Therefore, in Embodiment 4, a method for reducing the computational load will be described.

[0064] FIG. 10 is a diagram for explaining a processing method for reducing the computational load. At the lower part of the figure, it shows a state where an intersection where a right turn is planned is approaching and the determination point Dp is being approached. In this state, the driving assistance device 10 has decided to intensively monitor the right front area where it will turn right, and this is achieved by omitting the monitoring of other areas (for example, the left side and the left front). For example, when using LiDAR as a sensor, regarding the information of the target object measurement unit 18 obtained from the output of sensor 3, the measurement range of the LiDAR is specialized for the target object D2a in the right front. This can reduce the load required for the processing of the target object measurement unit 18.

[0065] At the upper part of FIG. 10, when using a camera as a sensor, it schematically shows an image grasped by the sensor, indicating that image processing of the right front area surrounded by a square is emphasized.

[0066] In this way, when approaching the position of the determination point Dp, the computational load can be reduced by performing recognition processing focusing on the target object to be measured. Alternatively, by allocating time to the recognition processing of the target object to be measured, it becomes easier to recognize the target object to be measured, and there is an effect that the stop of driving assistance / automatic driving is reduced.

[0067] In this case, the target object measurement unit 18 preferably uses the determination point and, when approaching the position of the determination point, performs external recognition processing based on the position of the target object to be measured.

Embodiment

[0068] In Embodiment 5, measures are taken for the problem of occlusion where the object behind that should originally be measured becomes invisible due to the object in front.

[0069] FIG. 12 illustrates the occlusion state. When the host vehicle 1 approaches an intersection and enters the stage of making a confirmation at the determination point Dp, there is another vehicle 1X at a position blocking the measurement target feature D2a, indicating that the measurement target feature D2a is not being measured.

[0070] FIG. 11 is a diagram showing a configuration example of the driving support device according to Embodiment 5. At this time, the driving support device 10 acts on the trajectory planning unit 14 to perform processing. First, the driving support state setting unit 13 determines that the reason why the measurement target feature D2a at the determination point Dp cannot be measured is occlusion by a moving object (another vehicle 1X), and transmits this to the trajectory planning unit 14. The trajectory planning unit 14 plans the speed of the host vehicle from the distance to the position of the determination point Dp and the speed of the moving object (another vehicle 1X) so that the occlusion is resolved by the time it reaches the determination point Dp, and drives accordingly.

[0071] In this way, the driving support state setting unit recognizes that the reason for being unable to recognize the measurement target feature is occlusion by a moving object, and the trajectory planning unit 14 plans the speed of the host vehicle from the distance to the position of the determination point Dp and the speed of the moving object so that the occlusion by the moving object is resolved by the time it reaches the position of the determination point Dp.

[0072] As a result, the occlusion is eliminated at the determination point Dp, and the measurement target feature D2a becomes measurable. According to Embodiment 5, there is an effect that the suspension of driving support / automatic driving is reduced.

Explanation of Reference Numerals

[0073] 1: Vehicle, 10: Driving support device, 11: Trajectory creation processing unit, 12: Trajectory driving confirmation processing unit, 13: Driving support continuation determination unit, 14: Trajectory planning unit, 15: Determination point setting unit, 16: Trajectory generation unit, 17; Sensor, 18: Target feature measurement unit, 19: Self-position estimation unit

Claims

1. A determination point setting unit that sets a determination point associating a hypothetical self-position of a host vehicle assumed on a map, a position on the map of a measurement target feature that can be measured from the hypothetical self-position, and an error predicted as a difference between the position on the map of the measurement target feature and an actual position; An object feature measurement unit that obtains a measurement position of the measurement target feature based on external information acquired by an external sensor mounted on the host vehicle; A self-position estimation unit that estimates a self-position of the host vehicle on the map based on the external information; A driving support state setting unit that sets a control mode of a driving support system of the host vehicle based on the determination point, the measurement position of the measurement target feature measured by the object feature measurement unit, and the self-position of the host vehicle estimated by the self-position estimation unit. A driving support device characterized by comprising:

2. The driving support device according to claim 1, further comprising an orbit generation unit that generates a travel orbit of the host vehicle starting from the hypothetical self-position based on the hypothetical self-position on the map, the position and error of the measurement target feature on the map; The determination point setting unit sets, as a determination point, the hypothetical self-position as the starting point of the travel orbit and the position and error of the measurement target feature on the map that can be measured from the hypothetical self-position when the orbit generation unit cannot generate the travel orbit that satisfies the vehicle motion allowance of the host vehicle. A driving support device characterized by:

3. The driving support device according to claim 1, The determination point setting unit corrects the position of the determination point forward based on at least one of a steering angle of the host vehicle at the hypothetical self-position, a change in a road area on the map, and a reaction time of a driver. A driving support device characterized by:

4. The driving support device according to claim 2 or claim 3, The determination point setting unit adds a new determination point notified as a transfer notice to the driver to a position before a preset time of the determination point, and the determination point includes information on a control mode of driving. A driving support device characterized by:

5. The driving support device according to claim 2, When the determination point setting unit allows only forward movement as the vehicle movement allowance and does not allow reverse movement, the control mode is set to inoperable for continuous driving and driver transfer, which is a feature of the driving support device.

6. The driving support device according to claim 2, wherein the determination point setting unit uses the turning radius of the host vehicle as the vehicle movement allowance, and sets the upper limit speed of the driving support system to be smaller as the turning radius is smaller, which is a feature of the driving support device.

7. The driving support device according to claim 1, wherein after the self-position and the assumed self-position match, when the measurement target feature corresponding to the assumed self-position cannot be recognized, the driving support device is characterized in that it is determined that the automatic driving by the driving support system of the host vehicle cannot continue.

8. The driving support device according to claim 2, wherein the determination point setting unit sets the determination point for the measurement target feature on the map existing on the left side of the self-position, which is a feature of the driving support device.

9. The driving support device according to claim 2, wherein the determination point setting unit sets the determination point based on the position of the assumed obstacle on the map instead of the position and error of the measurement target feature on the map, and after the self-position and the assumed self-position match, when it cannot be recognized that there is no assumed obstacle corresponding to the assumed self-position, the driving support device is characterized in that it is determined that the automatic driving by the driving support system of the host vehicle cannot continue.

10. The driving support device according to claim 2, wherein the determination point setting unit uses the measurement position of the measurement target feature measured by the target feature measurement unit, compares the measurement position with the position of the measurement target feature on the map, and sets the magnitude of the error of other measurement target features existing around the host vehicle, which is a feature of the driving support device.

11. The driving support device according to claim 1, wherein when approaching the position of the determination point using the determination point, external recognition processing is performed based on the position of the measurement target feature, which is a feature of the driving support device.

12. The driving support device according to claim 1, comprising a trajectory planning unit that plans a travel trajectory by the driving support system of the host vehicle. The driving support state setting unit recognizes that the reason for being unable to recognize the measurement target feature is occlusion by a moving object, The trajectory planning unit plans the speed of the host vehicle so that the occlusion by the moving object is resolved by the time the host vehicle reaches the position of the determination point, based on the distance to the position of the determination point and the speed of the moving object. A driving support device characterized by the above.

13. A driving support method in which a computer device generates a driving trajectory of a vehicle on a map and supports the driving of the vehicle along the driving trajectory, On the driving trajectory generated using the feature positions on the map, the position on the driving trajectory that enables stable operation when making a right or left turn according to a given vehicle movement allowance is set as a determination point in consideration of the feature error. A driving support method characterized by determining the availability of driving support according to the measured feature position when the measured vehicle position reaches the determination point during actual driving of the vehicle.

14. The driving support method according to claim 13, During actual driving of the vehicle, a difference is obtained between the measured feature position and the feature position on the map, and the difference is reflected in the driving trajectory as the feature error. A driving support method characterized by the above.

15. The driving support method according to claim 13, The driving trajectory is generated using obstacles assumed on the map. A driving support method characterized by the above.

Images