Automatic driving device and rule judgment device

Abstract

The present invention is an automatic driving device (100) used in a vehicle capable of automatic driving, comprising: a controllability judgment unit (173) that allows a change in the route of travel when the distance between the vehicle and surrounding vehicles existing around the vehicle is greater than a minimum control-allowed distance; and a minimum distance determination unit (172) that selects a relationship from relationships that determine the minimum control-allowed distance specified for each driving condition based on the driving condition of the vehicle, and uses the selected relationship to determine the minimum control-allowed distance used by the controllability judgment unit (173).

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application is based on Japanese Patent Application No. 2019-228547 filed in Japan on December 18, 2019, the contents of which are incorporated herein by reference in their entirety. Technical Field

[0003] The present invention relates to an automatic driving device, a rule judgment device, an automatic driving method and a rule judgment method. Background Art

[0004] It is known that there is a technology that enables vehicles to drive automatically. The device disclosed in Patent Document 1 calculates a safe distance. In addition, the device disclosed in Patent Document 1 calculates a potential accident liability value based on the safe distance. The potential accident liability value is a value that represents the degree of liability incurred by the vehicle in the event that an accident occurs between the vehicle and surrounding vehicles. The module for determining the potential accident liability value disclosed in Patent Document 1 calculates the potential accident liability value when driving on a path planned in the previous section, and if the calculated potential accident liability value is a value within a controllable range, control is allowed. The description of the technical elements in this specification is provided by referring to the contents of the prior art documents.

[0005] Patent Document 1: International Publication No. 2018 / 115963

[0006] The technology disclosed in Patent Document 1 is not a technology for changing the determination rule used in the automatic driving control according to the change of the driving condition. Therefore, a device or the like for determining whether the automatic driving control is appropriate in accordance with the change of the driving condition is required. Summary of the invention

[0007] The present disclosure is completed based on this situation, and its purpose is to provide an automatic driving device, a rule judgment device, an automatic driving method, and a rule judgment method that can determine whether it is appropriate automatic driving control corresponding to changes in driving conditions.

[0008] The above-mentioned object is achieved by the combination of features described in the independent claims, and the dependent claims provide more advantageous specific examples. The reference numerals in parentheses in the claims indicate the correspondence with the specific units described in the embodiments described later as one mode, and do not limit the disclosed technical scope.

[0009] A first disclosure of an automatic driving device for achieving the above-mentioned object is an automatic driving device used in a vehicle capable of automatic driving, comprising:

[0010] a controllability determination unit that allows a change in the travel path when the distance between the vehicle and a surrounding vehicle existing around the vehicle is greater than a minimum control-permitted distance; and

[0011] The minimum distance determination unit selects a relationship from relationships for determining the minimum control permission distance prescribed for each driving condition based on the driving condition of the vehicle, and determines the minimum control permission distance used by the controllability determination unit using the selected relationship.

[0012] The automatic driving device determines the control permission minimum distance used by the control determination unit for determination according to the driving condition in which the vehicle is traveling. Therefore, since the control permission minimum distance can be shortened in a driving condition where an adverse situation is unlikely to occur even if the inter-vehicle distance is shortened, it is possible to determine whether the automatic driving control is appropriate in response to the change in the driving condition.

[0013] The first disclosed automatic driving device includes a departure value determination unit. When a vehicle equipped with the automatic driving device, i.e., the vehicle itself, is traveling on a candidate path planned by a path planning unit, a departure value is determined for each candidate path, which indicates a possibility that the vehicle may deviate from the driving rules, based on a comparison result between the vehicle-to-vehicle distance between the vehicle and surrounding vehicles and a minimum distance allowed by control. The path planning unit sequentially plans candidate paths, which are candidates for the paths that the vehicle should travel next in order to continue traveling. The controllability judgment unit selects a path whose departure value determined by the departure value determination unit for each candidate path is within a control-allowed range from the candidate paths planned by the path planning unit as the path to be traveled by the vehicle, and outputs an instruction to travel on the selected path to the driving unit, which performs driving control for the vehicle to travel.

[0014] The driving conditions include specific route change conditions and standard driving conditions. In the specific route change conditions, there is a section where the route must be changed to a planned route. The planned route is a route that is longer than a candidate route and is the route taken by the vehicle when it reaches the destination. In the section where the route must be changed, the possibility that the vehicle-to-vehicle distance is longer than the minimum distance allowed by standard control determined based on a relationship specified for the standard driving conditions is lower than a threshold.

[0015] In a section where a lane change is necessary, a specific lane change status is defined based on whether the lane after the lane change is congested.

[0016] The minimum distance determination unit selects a relationship from relationships for determining the minimum control permission distance prescribed for each driving condition based on the current position of the vehicle, and determines the minimum control permission distance used by the controllability determination unit using the selected relationship.

[0017] The automatic driving device includes a rule acquisition unit that acquires a rule for determining a deviation value under a driving condition in which the host vehicle is driving, and a deviation value determination unit that determines the deviation value using the rule acquired by the rule acquisition unit.

[0018] The rule acquisition unit acquires a rule for determining a deviation value in a travel area in which the host vehicle is traveling, and acquires the rule in units of countries as travel areas.

[0019] The rule acquisition unit acquires a decision rule for the minimum control distance allowed for a driving condition from a server, and the server determines the decision rule for the minimum control distance allowed for each driving condition through learning.

[0020] The automatic driving device includes a notification control unit, and has a section where a route change must be made to a planned driving route. In the section where the route change must be made, when the possibility that the vehicle-to-vehicle distance is longer than the minimum distance allowed by control is lower than a notification threshold, the notification control unit outputs a notification from the notification unit indicating a high possibility that automatic control cannot be performed.

[0021] The automatic driving device includes a confirmation control unit, and has an interval in which a route change to a planned driving route must be made. In the interval in which the route change must be made, when the possibility that the vehicle-to-vehicle distance is longer than the minimum control distance allowed determined based on a standard relationship that determines the minimum control distance allowed is lower than a first threshold, but the possibility that the vehicle-to-vehicle distance is longer than the minimum control distance allowed determined by the minimum distance determination unit is higher than a second threshold set to be greater than the first threshold, the confirmation control unit confirms with the vehicle occupant whether to perform control to change the minimum control distance allowed to the minimum control distance allowed corresponding to the driving condition.

[0022] A second disclosure of an automatic driving device for achieving the above-mentioned object is an automatic driving device used in a vehicle capable of automatic driving, comprising:

[0023] a deviation value determination unit that determines, for each candidate path, a value indicating a possibility that the vehicle deviates from the driving rule, i.e., a deviation value, based on a comparison result of a vehicle distance between the vehicle and surrounding vehicles existing around the vehicle and a minimum control allowable distance when the vehicle equipped with the automatic driving device, i.e., the vehicle, is traveling on a candidate path planned by the path planning unit, wherein the path planning unit sequentially plans the candidate paths, the candidate paths being candidates for the paths that the vehicle should travel next in order to continue traveling;

[0024] a controllability determination unit that selects, from the candidate routes planned by the route planning unit, a route whose deviation value determined by the deviation value determination unit for each candidate route is within the control allowable range as a route for the host vehicle to travel, and outputs an instruction to travel on the selected route to the driving unit, and the driving unit performs driving control for the host vehicle to travel; and

[0025] The rule updating unit updates the rule used by the deviation value determining unit to determine the deviation value based on the changed sensor when a sensor detecting the behavior of at least one of the surrounding vehicles and the host vehicle is changed.

[0026] In the present automatic driving device, when the sensor is changed, the rule updating unit updates the rule for determining the deviation value. Thus, the deviation value reflecting the fact that the sensor has been changed can be determined. Furthermore, the controllability determination unit selects a path based on the deviation value determined by the updated rule. Therefore, when the sensor is changed, it is possible to determine whether the vehicle is traveling under automatic driving control appropriate for the path.

[0027] A third disclosure of an automatic driving device for achieving the above-mentioned object is an automatic driving device used in a vehicle capable of automatic driving, comprising:

[0028] a rule determination unit for determining whether the vehicle, which is a vehicle equipped with an automatic driving device, is following a driving rule when the vehicle is traveling on a candidate path planned by the path planning unit, wherein the path planning unit sequentially plans the candidate paths, which are candidates for paths that the vehicle should travel next in order to continue traveling;

[0029] a rule management unit that manages driving rules so that driving rules corresponding to driving conditions are applied as driving rules used for determination by the rule determination unit; and

[0030] The rule changing unit changes the driving rule used by the rule determining unit for determination according to the driving situation.

[0031] The disclosure of a rule judgment device for achieving the above-mentioned purpose is a rule judgment device having:

[0032] a rule determination unit for determining whether the vehicle, which is a vehicle equipped with an automatic driving device, is following a driving rule when the vehicle is traveling on a candidate path planned by the path planning unit, wherein the path planning unit sequentially plans the candidate paths, which are candidates for paths that the vehicle should travel next in order to continue traveling; and

[0033] The rule changing unit changes the driving rule used by the rule determining unit for determination according to the driving situation.

[0034] The automatic driving device and the rule judgment device can apply the driving rules corresponding to the driving conditions of the vehicle to judge whether the vehicle complies with the driving rules. Therefore, it can judge whether the automatic driving control is appropriate corresponding to the change of the driving conditions.

[0035] The first disclosure of the automatic driving method for achieving the above-mentioned object corresponds to the first disclosure of the automatic driving device. That is, the first disclosure of the automatic driving method is an automatic driving method used in a vehicle capable of automatic driving, wherein:

[0036] When the distance between the vehicle and surrounding vehicles is greater than the minimum distance allowed by the control, the lane change is allowed.

[0037] Based on the driving condition of the vehicle, one relationship is selected from relationships for determining the minimum control permission distance prescribed for each driving condition, and the selected relationship is used to determine the minimum control permission distance for determining whether to permit a lane change.

[0038] The second disclosure of the automatic driving method for achieving the above-mentioned object corresponds to the second disclosure of the automatic driving device. That is, the second disclosure of the automatic driving method is an automatic driving method used in a vehicle capable of automatic driving, wherein:

[0039] The candidate routes are planned in sequence. The candidate routes are candidates for the routes that the vehicle using the automatic driving method, i.e., the vehicle itself, should travel next in order to continue driving.

[0040] When the vehicle is traveling on the planned candidate route, a deviation value, which indicates the possibility of the vehicle deviating from the driving rules, is determined for each candidate route based on the comparison result between the inter-vehicle distance between the vehicle and the surrounding vehicles existing around the vehicle and the minimum distance allowed by the control.

[0041] Performing driving control for the vehicle to travel,

[0042] A path whose deviation value determined for each candidate path is within the control allowable range is selected from the planned candidate paths as the path for the vehicle to travel, and an instruction to travel on the selected path is output for the vehicle to travel.

[0043] When a sensor that detects the behavior of at least one of the surrounding vehicles and the host vehicle is changed, the rule for determining the deviation value is updated based on the changed sensor.

[0044] The disclosure of the rule judgment method for achieving the above-mentioned purpose corresponds to the disclosure of the above-mentioned rule judgment device. That is, the disclosure of the rule judgment method is a rule judgment method having:

[0045] When the vehicle equipped with the automatic driving device, namely the vehicle itself, is traveling on a candidate route, determining whether the vehicle is following the driving rules, wherein the candidate route is a candidate for a route that the vehicle should travel next in order to continue traveling; and

[0046] The driving rules used for judgment are changed according to the driving conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a diagram showing the structure of the automatic driving device 100 according to the first embodiment.

[0048] Figure 2 This is a diagram showing a part of the processing executed by the path planning unit 160 .

[0049] Figure 3 It means judging each candidate path T i Is it a cautious path? safe Processing diagram.

[0050] Figure 4 It is a diagram showing the structure of an automatic driving device 200 according to the second embodiment.

[0051] Figure 5 This is a diagram showing the processing executed by the path determination unit 176 and the notification control unit 177.

[0052] Figure 6 It is a diagram showing the structure of an automatic driving device 300 according to a third embodiment.

[0053] Figure 7 It is a diagram showing the processing executed by the path determination unit 176 and the confirmation control unit 178.

[0054] Figure 8 It is a diagram showing the structure of an automatic driving device 400 according to a fourth embodiment.

[0055] Fig. 9 It is a diagram showing the structure of an automatic driving device 500 according to the fifth embodiment.

[0056] Fig.10 It is a diagram showing the structure of an automatic driving device 600 according to the sixth embodiment.

[0057] Fig.11 It is a diagram showing the structure of an automatic driving device 700 according to the seventh embodiment.

[0058] Fig.12 Yes means Fig.11 A diagram showing a portion of the processing performed by the basic value determination unit 773.

[0059] Fig.13 It is a diagram showing the structure of an automatic driving device 800 according to the eighth embodiment.

[0060] Fig.14 It is a diagram showing the structure of an automatic driving device 900 according to a ninth embodiment.

[0061] Fig.15This is a diagram showing the processing executed by the path determination unit 176 and the confirmation control unit 178 in the second modification. DETAILED DESCRIPTION

[0062] <First embodiment>

[0063] The following is an explanation of the implementation method based on the accompanying drawings. If the behavior of manually driving a vehicle is considered, there is a case where the driver changes the inter-vehicle distance according to the driving area. For example, when the road at the merging destination is crowded at the merging point, there is a case where the inter-vehicle distance is narrower than the usual inter-vehicle distance when moving to the road at the merging destination. This is because if the inter-vehicle distance is not narrower than the usual inter-vehicle distance, there is a case where it is impossible to enter the road at the merging destination.

[0064] The driver of the vehicle traveling on the road at the merging destination is also aware of the concern that if the road is congested, vehicles traveling on the other merging road may enter the road at the merging destination with a narrow vehicle distance. Therefore, vehicles traveling on the other merging road are also allowed to enter the road at the merging destination with a narrow vehicle distance. In other words, it can be said that the vicinity of the merging point is an area where the vehicle distance can be narrow.

[0065] When controlling the vehicle-to-vehicle distance that allows lane change control by taking into account the minimum distance (hereinafter referred to as the control allowed minimum distance), if the special situation near the above-mentioned merging point is not considered, there is a concern that the vehicle in automatic driving may not be able to enter the road at the merging destination and may have to stop.

[0066] In addition, when changing lanes to a congested lane, if it is a manually driven vehicle, after indicating the intention to change lanes by flashing the direction indicator, the lane change is slowly performed, so that other vehicles traveling in the lane of the lane change destination can clearly identify the vehicle that is changing lanes. Other vehicles traveling in the lane of the lane change destination help the merging of the vehicle that is changing lanes by slowing down for the vehicle that is changing lanes. Therefore, in such a situation, even if the distance between vehicles is narrow, it is difficult to cause an adverse situation. In this way, depending on the driving conditions, even if the distance between vehicles is narrow, it is difficult to cause an adverse situation. In addition, the driving area is also one of the driving conditions.

[0067] If the deviation value, which is a value indicating the possibility of the vehicle deviating from the driving rules and determined by considering the minimum control distance, does not take into account the special situation of changing the lane to the congested road mentioned above, the vehicle in automatic driving may not be able to change the lane. The automatic driving device 100 disclosed in the first embodiment is a device that takes this situation into consideration.

[0068] Figure 11 is a diagram showing a configuration of an automatic driving device 100 according to the first embodiment. The automatic driving device 100 is mounted on a host vehicle 1. The host vehicle 1 is a vehicle that is mounted with the automatic driving device 100 when a certain automatic driving device 100 is used as a reference.

[0069] The vehicle is not particularly limited as long as it is a vehicle traveling on the road. Vehicles include ordinary passenger cars, trucks, buses, etc. The automatic driving device 100 is a device that controls the behavior of the vehicle 1. The behavior includes speed and direction of travel. The automatic driving method performed by the automatic driving device 100 is a vehicle control corresponding to an automatic driving level of level 3 or above.

[0070] The automatic driving device 100 includes one or more sensors 101. The sensor 101 is a sensor for detecting the behavior of surrounding vehicles and outputting sensor values ​​representing the behavior of surrounding vehicles. The sensor 101 may include a camera. In addition, the sensor 101 may also include a millimeter wave radar or LIDAR. Figure 1 In FIG. 1 , as the sensor 101 , sensors 101 a , 101 b , and 101 c are shown. Figure 1 The number of sensors 101 shown is an example, and the number of sensors 101 is not particularly limited.

[0071] The sensor 101 also includes sensors for detecting the position of the vehicle 1 and the behavior of the vehicle 1. If the current position of the vehicle 1 (hereinafter referred to as the vehicle position) can be detected in sequence, the speed and direction of travel of the vehicle 1 as the behavior of the vehicle 1 can be determined. Therefore, a sensor for detecting the position of the vehicle may be provided instead of the sensor 101 for directly detecting the behavior of the vehicle 1. The sensor 101 for detecting the position of the vehicle may also include a GNSS receiver. The sensor 101 for detecting the behavior of the vehicle 1 may include a vehicle speed sensor, a yaw rate sensor, an acceleration sensor, and the like.

[0072] In addition, the current position of the vehicle 1 can be detected by comparing the shape of the surroundings of the vehicle 1 detected by LIDAR, etc. with a high-precision map. In this case, the position of the vehicle 1 and the behavior of the vehicle 1 can be detected by using the sensor 101 that detects the behavior of surrounding vehicles, without having a dedicated sensor 101 for detecting the position of the vehicle 1 and the behavior of the vehicle 1.

[0073] The map storage unit 120 is a storage unit that stores a digital road map. The digital road map may be the high-precision map described above, or may be a normal road map that is not a high-precision map. A high-precision map is a map that also shows the positions of road signs such as lanes on the road, the types and positions of road signs, and three-dimensional objects around the road. In this specification, the storage unit includes a tangible storage medium that can be read by a computer. As the storage medium, for example, a flash memory can be used.

[0074] The rule DB storage unit 130 is a storage unit that stores a rule database (hereinafter, rule DB). The rule DB is a database that stores driving rules for each driving area. The driving rules for each driving area include the direction of travel such as one-way streets, speed limits, priority and non-priority distinctions, etc.

[0075] The driving rules are data that can be referred to in the judgment process of the controllability judgment unit 173. The rule DB storage unit 130 has a storage medium such as RAM or ROM. In the case where the judgment process of the controllability judgment unit 173 is implemented using a learned model including a neural network, the driving rules can also be given as teacher data used when learning the learned model.

[0076] The driving rules may include at least one of longitudinal speed rules, lateral speed rules, driving direction priority rules, signal light-based rules, traffic sign-based rules, and route priority rules. The longitudinal direction is the direction along the road or the front-rear direction of the vehicle, and the lateral direction is the width direction of the road or the width direction of the vehicle.

[0077] The driving rules may include in part or in whole the traffic rules of the region to overlap with part or all of the rules listed in the preceding text or as another rule. The driving rules may also be an element for installing a driving strategy. The driving strategy may also be defined as the strategy and rules that define the control actions in the vehicle level. The driving strategy may also be defined as the installation of the meaning of the vehicle level safety strategy (VLSS) to determine the level. The driving strategy may also be defined as the mapping from the detection state to the driving command.

[0078] The wireless communication unit 140 is a part that performs wireless communication with the outside of the vehicle 1, and includes an antenna for receiving and transmitting radio waves and a communication circuit that demodulates the radio waves received by the antenna and modulates the signals sent from the antenna. The wireless communication unit 140 performs wireless communication with the server 2 installed outside the vehicle 1. The wireless communication method may be a wide area wireless communication method via a base station and a public communication line network.

[0079] The sensor integration unit 150, the path planning unit 160, the rule determination unit 170, and the driving unit 180 can be implemented by a structure having at least one processor. For example, the sensor integration unit 150, the path planning unit 160, the rule determination unit 170, and the driving unit 180 can be implemented by a computer having at least one processor, a non-volatile memory, a RAM, an I / O, and a bus connecting these structures. In the non-volatile memory, a program for making a general-purpose computer function as the sensor integration unit 150, the path planning unit 160, the rule determination unit 170, and the driving unit 180 is stored. The processor uses the temporary storage function of the RAM and executes the program stored in the non-volatile memory, so that the computer functions as the sensor integration unit 150, the path planning unit 160, the rule determination unit 170, and the driving unit 180. Executing these functions means executing the vehicle control method corresponding to the program.

[0080] The sensor integration unit 150, the route planning unit 160, the rule determination unit 170, and the driving unit 180 can be implemented by different processors, respectively. Alternatively, the sensor integration unit 150, the route planning unit 160, the rule determination unit 170, and the driving unit 180 can be implemented by a structure having three or less processors.

[0081] The sensor value is input from the sensor 101 to the sensor integration unit 150. The sensor integration unit 150 includes a target vehicle behavior determination unit 151, a position acquisition unit 152, an environment determination unit 153, and a rule acquisition unit 154.

[0082] The target vehicle behavior determination unit 151 acquires the sensor value from the sensor unit 110. Then, based on the acquired sensor value, the relative behavior V of the target vehicle is sequentially determined. state .

[0083] The target vehicle is a vehicle selected from the surrounding vehicles existing in the vicinity of the vehicle 1. Whether a certain vehicle exists in the vicinity of the vehicle 1 can be determined, for example, by whether the vehicle is located in a surrounding area defined based on the vehicle 1. The surrounding area can be a rectangular area centered on the vehicle 1 and having sides parallel to the front-rear direction and the left-right direction of the vehicle. The size of the rectangle can be such that the front direction of the vehicle is used as the stopping distance of the vehicle. The rear of the vehicle can be the same as the front direction of the vehicle, or it can be shorter than the front direction of the vehicle. The size of the rectangle in the left-right direction of the vehicle can be the length of one lane. In addition, the size of the surrounding area can be set in various ways. In addition, the shape of the surrounding area can also be set in various ways. For example, the shape of the surrounding area can also be a perfect circle or an ellipse.

[0084] The target vehicle is a vehicle other than the vehicle 1 (hereinafter, other vehicle) existing in the surrounding area and having no other vehicle between the vehicle 1 and the target vehicle. In addition, the target vehicle may also be a vehicle existing in the surrounding area as long as the other vehicle exists between the vehicle 1 and the target vehicle.

[0085] Relative Behavior V state Including relative position and relative speed. Relative position can be expressed by relative distance and relative direction. Relative behavior V state The relative behavior V of the target vehicle can also be determined based on the change in the position of the host vehicle 1 and the position of the target vehicle. The position of the target vehicle is detected by the sensor 101 mounted on the target vehicle, or the automatic driving device 100 of the host vehicle 1 can obtain the position through wireless communication. In this case, the target vehicle behavior determination unit 151 obtains the sensor value detected by the sensor 101 mounted on the target vehicle, and determines the relative behavior V of the target vehicle based on the sensor value. state .

[0086] The position acquisition unit 152 acquires sensor values, and sequentially determines the current position of the host vehicle 1 , that is, the host vehicle position, based on the acquired sensor values.

[0087] The environment determination unit 153 determines the environment around the host vehicle 1 based on the sensor value. The environment includes the road shape of the road on which the host vehicle 1 is traveling. The road shape includes the road width, road inclination, road curvature, the shape of the road dividing line, etc. The surrounding environment of the host vehicle 1 may include information on obstacles around the host vehicle 1 other than surrounding vehicles in addition to the road conditions.

[0088] The rule acquisition unit 154 acquires the driving rules at the current vehicle position from the rule DB storage unit 130. The driving rules include rules for each driving area and rules that are independent of the driving area. The rules are used to determine the deviation value AL. val rules.

[0089] Used to determine the escape value AL val The rules include traffic rules. Below, there are the rules used to determine the escape value AL val The rules are only recorded as rules. Traffic rules are defined by laws and regulations. In order to obtain traffic rules for each driving area, the rule acquisition unit 154 determines the position of the vehicle 1 on the road based on the vehicle position defined by the sensor value and the road map stored in the map storage unit 120. In addition, the traffic rules defined based on the determined position on the road are acquired from the rule DB storage unit 130. The traffic rules acquired from the rule DB storage unit 130 include traffic rules around the vehicle position including the vehicle position (hereinafter, surrounding traffic rules Rdb).

[0090] The reason for acquiring the surrounding traffic rules Rdb is to prevent frequent acquisition of traffic rules from the rule DB storage unit 130 every time the vehicle position changes. An example of the surrounding traffic rules Rdb is the speed limit of the road at the vehicle position and its surroundings. In addition, the traffic rules acquired by the rule acquisition unit 154 from the rule DB storage unit 130 also include traffic rules for each country. There are cases where the traffic rules are different for each country, such as the meaning of road division lines being different for each country. Therefore, the rule acquisition unit 154 also acquires traffic rules for each country.

[0091] The rules that are not dependent on the driving area include, for example, rules that specify whether the vehicle can drive or stop according to the color of the traffic light. The rules that are not dependent on the driving area can also be stored in a specified storage area of ​​the rule DB storage unit 130. In the following, the traffic rules of the vehicle position and its surroundings and the rules that are not dependent on the driving area are combined as the surrounding traffic rules Rdb.

[0092] In addition, the rules include rules for maintaining the distance between vehicles above the minimum control distance allowed. It can be said that the minimum control distance allowed is a rule for the distance between vehicles that must be maintained. The minimum control distance allowed can be defined in two directions: the longitudinal minimum control distance allowed and the lateral minimum control distance allowed of the vehicle 1. The minimum control distance allowed can be simply set to the distance required for the vehicle 1 not to get too close to obstacles such as other vehicles. However, it is not simply a distance that is too close, but a distance that includes the margin for the vehicle 1 and other vehicles to continue driving in response to changes in the behavior of other vehicles around each vehicle.

[0093] If the actual inter-vehicle distance of the host vehicle 1 is less than the minimum control distance, the automatic driving device 100 controls the host vehicle 1 so as to immediately make the inter-vehicle distance greater than the minimum control distance if the automatic driving is in progress. Therefore, the automatic driving device 100 does not allow vehicle control when the inter-vehicle distance is less than the minimum control distance.

[0094] The minimum control distance is not always constant. The minimum control distance may vary according to the speed of the vehicle 1 and the speed of the surrounding vehicles. In addition, the minimum control distance varies according to the driving area. As a driving area, a merging area including a merging point is specified. For the merging area, the minimum control distance is set shorter than other driving areas. In addition, the driving area is not only a relatively narrow area such as a merging area, but also a relatively large area such as a country. The rule acquisition unit 154 acquires a rule for determining the minimum control distance in the driving area specified based on the position of the vehicle.

[0095] In addition, the rule acquisition unit 154 periodically acquires rules from the server 2 via the wireless communication unit 140, and stores the acquired rules in the rule DB storage unit 130. The server 2 determines the decision rule of the minimum control distance for each driving area by learning. The rule acquisition unit 154 acquires the decision rule of the minimum control distance for each driving area learned by the server 2.

[0096] The sensor integration unit 150 outputs the sensor basic information S to the path planning unit 160 and the rule determination unit 170 respectively. The sensor basic information S is the sensor value input to the sensor integration unit 150 and the information that can be derived from the sensor value. The information that can be derived from the sensor value includes the target vehicle information, the position of the vehicle, and the driving rules. The target vehicle information is the relative behavior V of the surrounding vehicles determined by the target vehicle behavior determination unit 151 based on the sensor value. state , information on the location and behavior of the vehicle 1.

[0097] The route planning unit 160 sequentially plans candidate routes T that are candidates for the short-term route that the host vehicle 1 should travel next. i (i=1, 2, 3...). The short-term path is a path for determining the control executed by the driving unit 180. The driving unit 180 controls the acceleration / deceleration and the travel direction of the host vehicle 1. The short-term path is a path that specifies the direction and speed at which the host vehicle 1 will travel in the next control cycle executed by the driving unit 180. The short-term path also includes time information, which specifies where the host vehicle 1 should be at a certain time.

[0098] The route planning unit 160 determines a candidate route T based on the relative behavior of the surrounding vehicles identified by the target vehicle behavior determination unit 151 and the surrounding environment of the host vehicle 1 determined by the environment determination unit 153. i As mentioned above, the candidate path T i It is a path that is a candidate for a short-term path. A short-term path is a path formed by dividing a long-term path into multiple paths, and is a path that can be traveled on the long-term path while avoiding surrounding vehicles. Therefore, a short-term path is a path for continuing to travel. A long-term path is a path from the current position to the destination. With respect to the destination, if the destination is set by the occupant of the vehicle 1, the destination can be set. In addition, the destination can also be set as the location where the vehicle 1 has traveled a certain distance on the road currently being traveled.

[0099] The path planning unit 160 plans a plurality of candidate paths T i . Alternative path T iThe path is a candidate for the short-term path instructed to the driving unit 180. Even if two paths are changed in lane to avoid the vehicle ahead, they are different candidate paths if the timing of the lane change is different. i As mentioned above, since the short-term path also includes time information, for example, even if the same straight-line path arrives at a different position after Δt seconds, it is a different candidate path T. i For the candidate path T planned by the path planning unit 160 i The number of candidate paths T planned by the path planning unit 160 is not particularly limited. i The number of can be changed according to the situation. The path planning unit 160 plans the candidate path T i Send to the rule determination unit 170.

[0100] The rule determination unit 170 includes a deviation value determination unit 171, a controllability determination unit 173, a deviation value storage unit 174, and an output unit 175. i When the vehicle 1 is traveling on the road, the deviation value determination unit 171 predicts the possibility that the vehicle 1 deviates from the driving rules. This possibility is expressed by the deviation value AL val To indicate the escape value AL val If the vehicle is between the target vehicle and the host vehicle 1, the value indicating the possibility of deviation from the driving rules is relatively large compared with the target vehicle. When the possibility of deviation from the driving rules is high, it can be predicted that the vehicle has deviated from the driving rules to a greater extent. Therefore, it can be said that the possibility of deviation from the driving rules is a value indicating the degree of deviation from the driving rules. In the present embodiment, the smaller the degree of deviation from the driving rules, the greater the deviation value AL. val The smaller the value, the smaller the value. val The vehicle 1 is on the candidate path T i When the vehicle 1 is traveling on the road, it is determined whether the vehicle 1 complies with the value of the driving regulations.

[0101] Use the vehicle 1 to i The separation value AL is determined based on the comparison result between the inter-vehicle distance between the host vehicle 1 and the surrounding vehicles when traveling on the upper side and the control allowable minimum distance as one of the factors. val Since the control-allowed minimum distance varies depending on the driving area and the environment, it is necessary to sequentially determine the control-allowed minimum distance. Therefore, the departure value determination unit 171 includes a minimum distance determination unit 172 that determines the control-allowed minimum distance.

[0102] The minimum distance determination unit 172 determines the control-permitted minimum distance using a relationship for determining the control-permitted minimum distance specified for each travel area. The rule acquisition unit 154 acquires the relationship from the rule DB storage unit 130 .

[0103] The above relationship is a relationship for defining the minimum control distance based on the speed of the vehicle that is the object of calculating the minimum control distance, and can be expressed by a mathematical formula, for example. The minimum control distance includes the minimum control distance with the preceding vehicle, that is, the longitudinal minimum control distance, and the lateral minimum control distance. The rule acquisition unit 154 acquires the relationship for determining these two minimum control distances.

[0104] The vehicle 1 is placed on the candidate path T i The separation value AL is determined by comparing the distance between the vehicle in front and the minimum distance allowed by longitudinal control. val In addition, as another factor, the vehicle 1 is on the candidate path T i The comparison result of the lateral vehicle distance when driving on the vehicle 1 and the lateral control allowed minimum distance. The lateral direction is either or both of the road width direction and the vehicle width direction of the vehicle 1. In addition, whether the path complies with the traffic regulations at the position of the vehicle is also an influence on the deviation value AL. val The value of the factor.

[0105] The escape value AL is preset based on these factors. val The escape value determination formula or the escape value determination map is hereinafter collectively referred to as the escape value determination relationship. The escape value determination unit 171 determines the escape value AL based on the escape value determination relationship and the above-mentioned factors. val .

[0106] The more the vehicle 1 drives in accordance with the driving rules, the smaller the deviation value AL val For example, when the inter-vehicle distance is sufficiently ensured, the deviation value AL val In addition, when the vehicle 1 accelerates suddenly, the separation value AL val There are cases where the value becomes large. This is because the minimum control permissible distance with the preceding vehicle becomes longer due to sudden acceleration.

[0107] Traffic regulations may also be acquired by other methods than acquiring from the regulation DB storage unit 130. Specifically, the traffic regulations at the current location may be acquired by analyzing images captured by a camera capturing the surroundings of the vehicle 1 and detecting signs, traffic lights, road markings, and the like.

[0108] The controllability determination unit 173 selects the candidate routes T planned by the route planning unit 160 from the candidate routes T i The selected route is referred to as the selected route T. passed . Select path T passed So be careful with the path T safe Condition. Careful Path T safe It is a path that does not deviate from the driving rules in relation to the target vehicle. passed Is a cautious path T safe In this case, it is allowed to select path T passed The automatic driving control on the road selects the path T passed Not a cautious path T safe In this case, autonomous driving control is not allowed.

[0109] If the deviation value AL determined by the deviation value determination unit 171 is val If the controllability determination unit 173 is within the controllable range, the deviation value AL val The candidate path T i As a cautious path T safe The control permission range is a deviation value AL which means that the vehicle 1 does not deviate from the driving rules. val The range includes 0. Alternatively, only 0 may be used as the control allowable range.

[0110] When the vehicle distance is shorter than the minimum distance allowed by the control, the disengagement value AL val Therefore, when the vehicle distance is shorter than the minimum distance allowed by the control, it is not allowed to be in the candidate path T i If the candidate path T is considered i In the case of a route that requires a course change, the inter-vehicle distance being greater than the control permission minimum distance becomes a condition for the control possibility determination unit 173 to permit the course change.

[0111] If it is not the value of AL val Alternative path T within the control allowable range i In the case of emergency stop, select the emergency stop path T e As the selected path T passed . Emergency stop path T e The rule determination unit 170 outputs the selected route T to the driving unit 180. passed .

[0112] The escape value storage unit 174 includes a writable non-volatile storage medium. passedStore the detached value AL together val The output unit 175 can be connected to an external device outside the automatic driving device 100. The output unit 175 has a signal transmission and reception function. The escape value AL stored in the escape value storage unit 174 val The signal is outputted to an external device connected to the output unit 175 via the output unit 175 .

[0113] The driving unit 180 selects a path T passed The driving mode determines the traveling direction and speed of the host vehicle 1. Then, based on the determined traveling direction and speed, the steering actuator provided in the host vehicle 1 and the driving force source and the braking device provided in the host vehicle 1 are controlled.

[0114] [Structure of Server 2]

[0115] The server 2 communicates with the wireless communication unit 140 mounted on various vehicles. In addition, the server 2 learns the determination rule of the control allowable minimum distance at various locations. As a method for learning the determination rule of the control allowable minimum distance, there is a method of acquiring the position from the probe car at any time. In addition, it is preferable to acquire not only the position but also the image of the surrounding area from the probe car. The probe car is a car having a function of sequentially sending the position, or the position and the image of the surrounding area of ​​the probe car to the server 2.

[0116] In addition, a roadside device that takes pictures of the driving area may be arranged in a predetermined driving area, and the server 2 may obtain the images taken by the roadside device. In this case, the server 2 analyzes the images obtained from the roadside device to determine the vehicle distance in the driving area.

[0117] In order to learn the decision rule of the minimum allowable distance, specifically, the relationship between the inter-vehicle distance, the vehicle's driving speed, and the number of vehicles per unit road length is learned. The number of vehicles per unit road length indicates the congestion of the road. Near the merging road, if the road becomes congested, there is a possibility that the lane change for merging is performed with a shorter inter-vehicle distance than the normal inter-vehicle distance specified according to the driving speed.

[0118] Even if the inter-vehicle distance is shorter than the normal inter-vehicle distance, if the lane change for merging is not performed, the vehicle traveling on the road to be merged cannot continue to travel. The driver of the vehicle traveling on the road to be merged and the driver of the vehicle traveling on the road to be merged are aware of this. Therefore, even if the lane change for merging is performed at an inter-vehicle distance shorter than the normal inter-vehicle distance, it is difficult to cause an adverse situation.

[0119] In other words, depending on the driving range, there is a case where the lane change is performed at a shorter inter-vehicle distance than the normal inter-vehicle distance specified by the driving speed, and it is possible to learn that even if such a lane change is performed, no adverse situation will occur. The inter-vehicle distance for lane change at this time is used as the minimum control allowable distance. When the number of vehicles per unit road length is how many, at what speed and inter-vehicle distance the lane change is performed, it is learned.

[0120] 〔To send selected path T passed Processing up to now

[0121] exist Figure 2 , it is shown that the path planning unit 160 will select the path T passed The process is performed until the signal is sent to the driving unit 180 . Figure 2 The processing shown is executed in each execution cycle. The execution cycle is set in advance. The execution cycle can be set arbitrarily, for example, it can be set to any time within the range of tens of milliseconds to hundreds of milliseconds.

[0122] In step (hereinafter, step omitted) S1, the sensor value is acquired from the sensor 101. This S1 is a process executed by the target vehicle behavior determination unit 151, the position acquisition unit 152, and the environment determination unit 153. In S1, various sensor values ​​required in the target vehicle behavior determination unit 151, the position acquisition unit 152, and the environment determination unit 153 are acquired.

[0123] S2 is executed by the target vehicle behavior determination unit 151. In S2, a target vehicle is selected from the surroundings of the host vehicle 1. The target vehicle selected here can be the same as the vehicle in the previous Figure 2 However, if a certain surrounding vehicle requires special attention compared to other surrounding vehicles, the frequency of selecting the surrounding vehicle requiring special attention as the target vehicle may be higher than that of the other surrounding vehicles. In S2, the sensor value obtained in S1 is further used to determine the relative behavior V of the target vehicle. state .

[0124] S3 is executed by the environment determination unit 153. In S3, the surrounding environment of the host vehicle 1 is determined using the sensor values ​​acquired in S1.

[0125] S4 is executed by the rule acquisition unit 154. In S4, the driving rules are acquired. As described above, the driving rules include rules that differ according to the driving area and rules that do not depend on the driving area. If the rules that do not depend on the driving area have been acquired, only the rules that differ according to the driving area are acquired.

[0126] As a rule that differs according to the travel area, there is a surrounding traffic rule Rdb. The surrounding traffic rule Rdb that is defined in the travel area based on the vehicle position included in the sensor value acquired in S1 is acquired from the rule DB storage unit 130 .

[0127] S5 and subsequent steps are executed by the deviation value determination unit 171. In S5, the deviation value determination unit 171 obtains the driving rule acquired by the rule acquisition unit 154 when executing S4, and determines the deviation value AL based on the acquired driving rule and the vehicle position. val Driving rules used when driving.

[0128] In S6, the candidate route T is obtained from the route planning unit 160. i In S7, the candidate path T obtained in S6 is determined. i Is there a cautious path T safe As mentioned above, the careful path T safe This is a path that does not deviate from the driving rules in relation to the target vehicle. Figure 3 The process shown in the figure determines each candidate path T i Is it a cautious path? safe . Figure 3 Processing shown.

[0129] In a cautious path T safe In the case of, enter S8. In S8, from the cautious path T safe The candidate path T i Choose the best path. When there is only one cautious path T safe In this case, the cautious path T safe As the best path. There are multiple cautious paths T safe In the case of, for example, the path that best matches the long-term path is taken as the optimal path.

[0130] In the judgment of S7, it is judged that there is no cautious path T safe In this case, enter S9. No careful path T safe It means that no matter which candidate path T is chosen i , the vehicle 1 may be responsible. In this case, an emergency stop is performed.

[0131] In S9, select emergency stop path T e The emergency stop path T of this embodiment e This is a path for causing the vehicle 1 to decelerate at the maximum speed and stop without performing steering control. passedis sent to the driving unit 180. In addition, in this S10, the selected path T passed The calculated escape value AL val and select path T passed The values ​​are stored as a group in the deviation value storage unit 174 .

[0132] 〔Is it a cautious path? safe Judgment of

[0133] exist Figure 3 In the figure, it indicates the judgment of each candidate path T i Is it a cautious path? safe In Figure 3 In the example, S11 to S17 are executed by the deviation value determination unit 171, and S18 and S19 are executed by the controllability determination unit 173. i , that is, whenever executing Figure 2 For S6, execute Figure 3 In addition, for each candidate path T i implement Figure 3 Processing shown.

[0134] In S11, it is estimated that in the candidate path T i The vehicle state St1 after Δt seconds when the vehicle is traveling on the emergency stop path T is inferred. The vehicle state St1 includes at least the position and speed of the vehicle 1. In S12, after the vehicle state St1 inferred in S11 is inferred, the vehicle state St1 inferred in the emergency stop path T is inferred. e The vehicle state St2 is a state when the vehicle 1 is stopped while traveling. Since the vehicle 1 is stopped, the vehicle state St2 mainly indicates the position of the vehicle 1.

[0135] In S13 , a target vehicle is selected from the surroundings of the host vehicle 1 . The target vehicle selected here is a vehicle different from the surrounding vehicles already selected as the target vehicle. In addition, the target vehicle is a vehicle selected from the surrounding vehicles existing around the host vehicle 1 .

[0136] In S14, an action set A that the target vehicle selected in S13 may take after Δt seconds is estimated. The action set A is one or more trajectories that the target vehicle may move within Δt seconds. The trajectory can be represented by the position of the target vehicle at multiple moments included in Δt seconds. Alternatively, the trajectory can also be represented by the latest position of the target vehicle at the time of executing S14 and the subsequent steering angle and speed.

[0137] The range of directions that the track may be directed to can be a certain range from the current traveling direction to the left and right sides based on the current traveling direction of the target vehicle. The certain range can also be a range that becomes narrower as the vehicle speed increases. The action set A of the target vehicle can be determined based on the target vehicle information.

[0138] In S15, for the action set A determined in S14, the deviation value AL during the state change of the vehicle 1 from the vehicle state St1 to the vehicle state St2 is calculated. val . At the departure value AL val The minimum control allowable distance is determined in the calculation of . The minimum control allowable distance is determined using one of the relationships selected from the relationships for determining the minimum control allowable distance prepared for each driving area according to the vehicle position.

[0139] In S16, for the action set A determined in S14, it is determined whether there is a problem in the driving of the vehicle 1 when the vehicle 1 changes from the vehicle state St1 to the vehicle state St2. val To determine whether there is a problem in the driving of the host vehicle 1 for the action set A. If the deviation value AL of the host vehicle 1 represented by the action set A relative to each track of the target vehicle val If it is a value indicating that a part of the host vehicle 1 deviates from the driving regulations, there is a problem in the driving of the host vehicle 1 .

[0140] According to the comparison result of the minimum distance allowed by the control and the vehicle distance, the candidate path T i Whether it complies with traffic regulations, etc., determines the departure value AL val In addition, for each candidate route, a deviation value AL is determined at each preset time interval. val .

[0141] In addition to the separation value AL val Except when the deviation value AL is 0, it is assumed that at least a part of the host vehicle 1 deviates from the driving rule. val Only when it is 0, it can be assumed that the vehicle 1 does not deviate from the driving regulations.

[0142] For each trajectory included in action set A, determine the departure value AL for action set A val . In determining multiple exit values ​​AL for an action set A val In the case of the maximum value of the escape value AL val To determine whether there is a problem in the driving of the vehicle 1.

[0143] If the result of S16 is "No", the process proceeds to S17. In S17, it is determined whether all vehicles that are the target vehicles have been checked for any problems in the driving of the vehicle 1. If the result of S17 is "No", the process returns to S13. If the result of S17 is "Yes", the process proceeds to S18. In S18, it is determined whether the candidate path T i Is a cautious path T safe .

[0144] If the result of S16 is "yes", the process proceeds to S19. In S19, it is determined that the candidate path T i Not a cautious path T safe Based on the careful path thus determined, T safe To judge S7, we need to know whether there is one or not.

[0145] [Summary of the first embodiment]

[0146] As described above, the automatic driving device 100 of the first embodiment determines the minimum control distance (S15) according to the driving area in which the host vehicle 1 is traveling. Since the minimum control distance corresponding to the driving area is determined, the minimum control distance can be shortened in driving areas such as merging roads where it is difficult to cause adverse conditions even if the vehicle is manually driven even if the inter-vehicle distance is shortened. By shortening the minimum control distance, it is possible to prevent the vehicle from being unable to drive and stopping during automatic driving, or from being unable to change the lane to the lane to be entered during automatic driving.

[0147] In the first embodiment, the control permission minimum distance and the vehicle distance are compared to determine the candidate route T for each candidate route T. i Calculate the detachment value AL val (S15). Then, the escape value AL val Alternative path T within the control allowable range i The path that the host vehicle 1 is to travel is indicated to the travel unit 180. The deviation value AL val is a value that relatively indicates the possibility that the vehicle 1 deviates from the driving rules between the vehicle 1 and the surrounding vehicles. val By traveling on a route within the control permission range, it is possible to suppress the conclusion that the driving of the host vehicle 1 is inappropriate due to a problem that occurs with surrounding vehicles during automatic driving.

[0148] The departure value determination unit 171 determines the departure value AL according to a rule corresponding to the driving area. valTherefore, it is possible to determine the deviation value AL based on the characteristics of each region, considering that, in a certain driving area, a situation that causes a problem in driving rules occurs as a result of a certain driving such as a change in the driving line, part of the cause of the situation lies in the vehicle 1, but there is no problem in the driving of the vehicle 1 in other driving areas. val As a difference in rules caused by such a difference in driving area, for example, there is a difference in the meaning of road dividing lines. In addition to rules stipulated by laws and regulations such as the meaning of road section lines, the difference in rules caused by the difference in driving area may also be a difference in rules formed based on experience.

[0149] In addition, in the present embodiment, the server 2 determines the determination rule of the control allowable minimum distance for each driving area by learning. Furthermore, the rule acquisition unit 154 acquires the determination rule of the control allowable minimum distance for each driving area learned by the server 2. Therefore, by determining the departure value AL according to the rule acquired by the rule acquisition unit 154 val , and according to the separation value AL val The route is selected so that the host vehicle 1 can continue traveling without stopping the host vehicle 1 excessively.

[0150] <Second embodiment>

[0151] Next, a second embodiment is described. In the second embodiment and the following description, except for the cases specifically mentioned, the elements with the same reference numerals as the reference numerals used previously are the same as the elements with the same reference numerals in the previous embodiments. In addition, when only a part of the structure is described, the previously described embodiment can be applied to the other parts of the structure.

[0152] exist Figure 4 2 shows a configuration of an automatic driving device 200 according to a second embodiment. The automatic driving device 200 is a configuration in which a path determination unit 176 , a notification control unit 177 , and a notification unit 182 are added to the automatic driving device 100 .

[0153] The rule determination unit 170 includes a route determination unit 176 and a notification control unit 177. The route determination unit 176 sequentially determines the route that the vehicle 1 will take when traveling to the destination, that is, the planned travel route. The planned travel route is the long-term route mentioned in the first embodiment. Therefore, the planned travel route is longer than the candidate route T. i The method for determining the planned travel route is the same as that for the long-term route. When the planned travel route is determined by an element different from the route determination unit 176 such as the route planning unit 160, the route determination unit 176 obtains the planned travel route from the element that determines the planned travel route.

[0154] When the planned travel route determined by route determination unit 176 includes a travel section where the possibility that the inter-vehicle distance is shorter than the control-permitted minimum distance exceeds the notification threshold, notification control unit 177 notifies notification unit 182 that automatic control is likely to be impossible.

[0155] The notification unit 182 is a configuration including one or both of a display screen and a speaker. The notification unit 182 is provided in the cabin of the host vehicle 1 at a position where the driver can recognize the notification from the notification unit 182 .

[0156] exist Figure 5 In FIG. 1 , the processing performed by the path determination unit 176 and the notification control unit 177 is represented as a flowchart. Figure 5 In the process, S21 is executed by the path determination unit 176 , and S22 to S26 are executed by the notification control unit 177 . Figure 5 The processing shown is repeatedly executed at a constant cycle while the host vehicle 1 is traveling under automatic driving.

[0157] In S21, a planned travel route is determined. In S22, traffic conditions when traveling on the planned travel route determined in S21 are acquired. The traffic conditions are acquired from the outside via the wireless communication unit 140, for example. The traffic conditions include traffic volumes classified by road sections and by time periods.

[0158] In S23, if the traffic conditions are the same as those obtained in S22, it is determined whether it is predicted that the automatic driving cannot be performed. The so-called automatic driving cannot be performed is a case where the traffic volume is heavy and the route change must be performed in the section, and the possibility that the inter-vehicle distance during the route change is longer than the control-allowed minimum distance is lower than the notification threshold. The control-allowed minimum distance uses the distance determined by the minimum distance determination unit 172.

[0159] In automated driving, if the probability that the distance between vehicles during a lane change is longer than the minimum distance allowed by control is low, it means that a lane change cannot be performed. In addition, lane changes here also include merging.

[0160] If the judgment result of S23 is "No", then end Figure 5 On the other hand, if the result of determination in S23 is "yes", the process proceeds to S24. In S24, the driver of the vehicle 1 is notified from the notification unit 182 that it is predicted that the automatic driving cannot be performed.

[0161] In S25, it is determined whether the driver has performed an operation to change to the manual driving mode. If the determination result of S25 is "No", the process ends. Figure 5 On the other hand, if the determination in S25 is "yes", the process proceeds to S26. In S26, the automatic driving is terminated.

[0162] According to the second embodiment, when it is predicted that the automatic driving cannot be performed (S23: Yes), the driver is notified of this in advance (S24). By making this notification, the driver can know in advance that the automatic driving cannot be performed. Therefore, it is possible to change to the manual driving mode at a timing that is easy for the driver to transfer the driving authority before the automatic driving cannot be performed.

[0163] <Third embodiment>

[0164] exist Figure 6 2 shows a configuration of an automatic driving device 300 according to a third embodiment. The automatic driving device 300 is a configuration in which the confirmation control unit 178 is added to the automatic driving device 200.

[0165] use Figure 7 The confirmation control unit 178 is described below. Figure 7 In FIG. 1 , the processing performed by the path determination unit 176 and the confirmation control unit 178 is shown as a flowchart. Figure 7 In the process, S31 is executed by the route determination unit 176, and S32 to S37 are executed by the confirmation control unit 178. During the period when the vehicle 1 is driving by automatic driving, the automatic driving control is repeatedly executed at a fixed cycle. Figure 7 Processing shown.

[0166] S31 is the same as S21, and determines the planned travel route. S32 is the same as S22, and acquires the traffic conditions when traveling on the planned travel route determined in S31.

[0167] In S33, in the case of the traffic conditions obtained in S32, it is determined whether there is an area where it is predicted that automatic driving cannot be performed under the standard control-allowed minimum distance. Specifically, in the case of the traffic conditions obtained in S32, it is determined whether the possibility that the vehicle distance when the route is changed is longer than the control-allowed minimum distance determined based on the standard relationship for determining the control-allowed minimum distance (i.e., the standard control-allowed minimum distance) is lower than a first threshold. The situation where the possibility that the vehicle distance when the route is changed is lower than the standard control-allowed minimum distance means that the route cannot be changed in automatic driving using the standard control-allowed minimum distance. The standard relationship for determining the control-allowed minimum distance is a relationship that can determine the control-allowed minimum distance without the particularity of the driving area.

[0168] If the judgment result of S33 is "No", then end Figure 7The process shown. On the other hand, if the judgment result of S33 is "yes", enter S34. In S34, if it is the minimum control distance corresponding to the driving area, it is determined whether automatic driving is possible. Specifically, in the case of the traffic conditions obtained in S32, it is determined whether the possibility that the vehicle distance when the route is changed is longer than the minimum control distance determined based on the relationship between the minimum control distance determined for each driving area (hereinafter, the minimum control distance divided by area) is higher than the second threshold. The second threshold is set to a value that is the same as the first threshold or a value that is greater than the first threshold.

[0169] If there is a possibility that the vehicle-to-vehicle distance when changing the route is longer than the minimum distance allowed by the control divided by area, it means that the route change can be performed if the automatic driving with the minimum distance allowed by the control divided by area is used. The minimum distance allowed by the control divided by area is determined by the minimum distance determination unit 172. If the judgment result of S34 is "No", it is predicted that automatic driving cannot be performed. Therefore, the execution Figure 5 S24 and beyond.

[0170] If the result of S34 is "yes", the process proceeds to S35. In S35, the occupant of the vehicle 1 is asked whether the automatic driving control of the minimum control distance is allowed to be changed. In S36, it is determined whether the occupant of the vehicle 1 allows the minimum control distance to be changed for each driving area. The occupant of the vehicle 1 can indicate whether the minimum control distance is allowed to be changed for each driving area by operating a predetermined button or the like.

[0171] If the judgment result of S36 is "No", then end Figure 7 On the other hand, if the determination result of S36 is "Yes", the process proceeds to S37. In S37, the automatic driving is performed using the relationship of the control permission minimum distance determined for each driving area.

[0172] [Summary of the Third Implementation Method]

[0173] The confirmation control unit 178 determines whether the possibility that the inter-vehicle distance is longer than the minimum distance allowed by the standard control is lower than a first threshold value (S33) in the section where the travel route must be changed in the scheduled travel route determined by the route determination unit 176. If the possibility is lower than the first threshold value, it is then determined whether the possibility that the inter-vehicle distance is longer than the minimum distance allowed by the control divided by area is higher than a second threshold value. If the possibility that the inter-vehicle distance is longer than the minimum distance allowed by the control divided by area is higher than the second threshold value, it is confirmed to the occupant of the vehicle 1 whether to perform control to change the minimum distance allowed by the control to the minimum distance allowed by the control divided by area (S35).

[0174] The control permissible minimum distance for each area is shorter than the standard control permissible minimum distance. Therefore, depending on the occupant, there is a possibility that the occupant may not like the automatic driving control using the control permissible minimum distance for each area. As in the present embodiment, by asking the occupant whether to allow the automatic driving control with the control permissible minimum distance changed, it is possible to suppress the execution of the automatic driving control that the occupant does not like.

[0175] <Fourth embodiment>

[0176] The autonomous vehicle uses the sensor group of the vehicle 1 to understand the surrounding conditions of the vehicle 1. If a change occurs in the sensor group, the acquired information changes. If the acquired information changes, even if the decision deviation value AL is changed before, val There is a possibility that the planned driving road may be considered to be out of the driving rules according to the rules, but it may also be considered to be not out of the driving rules.

[0177] Even if there is a change in the sensor group, as long as the decision escape value AL is not changed val Even if a change occurs in the sensor group, the departure value AL val The value will not be changed to reflect the change of the sensor group. In this case, appropriate autonomous driving control cannot be performed.

[0178] exist Figure 8 The structure of the automatic driving device 400 of the fourth embodiment is shown in FIG. The automatic driving device 400 is provided with a rule determination unit 470 instead of the rule determination unit 170 of the first embodiment. In addition, it is provided with an additional sensor unit 410 and a sensor integration unit 420. In addition, the automatic driving device 400 also includes a sensor 101, a map storage unit 120, a rule DB storage unit 130, a wireless communication unit 140, and a sensor integration unit 150, but for the sake of convenience in illustration, the following are omitted in FIG. Figure 8 These structures are omitted.

[0179] The additional sensor unit 410 is not initially disposed in the automatic driving device 400. After the automatic driving device 400 is mounted on the vehicle 1, the additional sensor unit 410 is added later in order to improve the automatic driving device 400. The additional sensor unit 410 is manufactured by a third party, for example.

[0180] The additional sensor unit 410 includes an additional sensor 411 and a control unit 412. The additional sensor 411 detects the separation value AL. val Such physical property values ​​include values ​​that determine the behavior of at least one of the surrounding vehicles or the own vehicle 1. The additional sensor 411 outputs the detected value to the sensor integration unit 420.

[0181] The control unit 412 controls the additional sensor 411 and provides the rule updating unit 471 with information for updating the decision deviation value AL. val The information of the rule is, for example, the type and installation location of the additional sensor 411. In addition, the control unit 412 may store the updated part of the rule and provide the stored updated part of the rule to the rule updating unit 471.

[0182] The sensor integration unit 420 integrates the sensor value detected by the additional sensor 411 into the sensor value detected by the sensor 101. For example, when the sensor 101 and the additional sensor 411 respectively detect the same behavior of the object, the behavior of the object detected by the sensor 101 and the behavior of the object detected by the additional sensor 411 are averaged, and one behavior is determined as the behavior of the object. The sensor integration unit 420 supplies the processing result to the deviation value determination unit 472 provided in the rule determination unit 470.

[0183] The rule determination unit 470 includes the controllability determination unit 173 described in the above embodiment, the deviating value storage unit 174, and the output unit 175. The rule determination unit 470 also includes a rule update unit 471, a deviating value determination unit 472, and a reference deviating value determination unit 473.

[0184] The sensor basic information S described in the above-mentioned embodiment is input to the rule determination unit 470 . The sensor basic information S is information that is not affected by the addition of the sensor unit 410 .

[0185] The rule updating unit 471 updates the deviating value AL determined by the deviating value determining unit 472 based on the information from the control unit 412 included in the additional sensor unit 410. val By updating the rule, the escape value determination unit 472 can determine the escape value AL reflecting the sensor value detected by the additional sensor 411. val The rule updating unit 471 is an example of a rule changing unit.

[0186] The deviated value determination unit 472 applies the rule updated by the rule updating unit 471 and determines the deviated value AL based on the information provided from the sensor integration unit 420. val The processing performed by the departure value determination unit 472 may be the same as the processing performed by the departure value determination unit 171, except that a part of the rule is different from that of the departure value determination unit 171 and the information used is information provided by the sensor integration unit 420. However, the departure value determination unit 472 may use the rule for determining the minimum control allowable distance as a rule that is not dependent on the driving area.

[0187] The reference escape value determination unit 473 determines the escape value AL using the sensor basic information S. val The reference deviation value determination unit 473 uses the sensor basic information S and determines the deviation value AL according to the rule before being updated by the rule updating unit 471. val In the present embodiment, the deviation value AL determined by the reference deviation value determination unit 473 is val The deviating value AL determined by the reference deviating value determining unit 473 is stored in the deviating value storing unit 174. val Therefore, the deviation value AL determined by the reference deviation value determination unit 473 is not provided to the control possibility determination unit 173. val To determine whether automatic control is possible.

[0188] The deviation value AL determined by the reference deviation value determination unit 473 is val The separation value AL is compared between the own vehicle 1 and other vehicles. val The reference separation value AL val Even if the additional sensor 411 is added, automatic control can be performed with higher accuracy, but when comparing the deviation value AL val When there is a separation value AL calculated according to the previously established rules val Therefore, in this embodiment, the deviated value AL stored in the deviated value storage unit 174 and used for comparison val The separation value AL val The escape value AL is determined by the rules before the change. val .

[0189] In the fourth embodiment, when the additional sensor unit 410 is added, the rule updating unit 471 updates the decision deviation value AL val Thus, the separation value AL reflecting the addition of the additional sensor 411 can be determined. val Then, the controllability determination unit 173 determines the escape value AL according to the updated rule. val To decide the path T passed Therefore, it is possible to perform the separation value AL according to which the reliability is improved by adding the additional sensor unit 410. val Highly reliable autonomous driving control.

[0190] In addition, in this embodiment, a reference deviation value determination unit 473 is also provided to determine a deviation value AL to which the rule before the update is applied. val , and the escape value AL val The deviating value AL is stored in the deviating value storage unit 174. val When creating a new rule, you can use the rules that have already been established.

[0191] <Fifth embodiment>

[0192] exist Fig. 9 2 shows the structure of the automatic driving device 500 of the fifth embodiment. The automatic driving device 500 does not include the reference deviation value determination unit 473 included in the automatic driving device 400. Instead, the deviation value AL determined by the deviation value determination unit 472 is used as the reference deviation value determination unit 473. val The deviation value is stored in the deviation value storage unit 174. The automatic driving device 500 is different from the automatic driving device 400 only in this point.

[0193] As in the automatic driving device 500, the deviation value AL stored in the deviation value storage unit 174 val It can also be the escape value AL determined by the updated rule val .

[0194] According to the fifth embodiment, during driving, the separation value determination unit 472 and the reference separation value determination unit 473 may not be sequentially executed to determine the separation value AL. val Therefore, the processing load is reduced.

[0195] <Sixth embodiment>

[0196] exist Fig.10 , the structure of the automatic driving device 600 of the sixth embodiment is shown. The automatic driving device 600 is provided with a rule updating unit 671 instead of the rule updating unit 471 provided in the automatic driving devices 400 and 500. In addition, the structure of the deviation value determining unit 672 provided in the rule judgment unit 670 is different from that of the deviation value determining unit 472. The deviation value determining unit 672 is provided with two detailed structures, namely, a basic value determining unit 673 and a correction unit 674. The rule updating unit 671 obtains the correction rule used by the correction unit 674. The rule updating unit 671 is an example of a rule changing unit.

[0197] The basic value determination unit 673 determines the deviation value AL based on the sensor basic information S. val Therefore, the basic value determination unit 673 is the same as the reference deviation value determination unit 473 of the fourth embodiment. The basic value determination unit 673 determines the deviation value AL val The rule does not take into account the value detected by the additional sensor 411. The basic value determination unit 673 determines the deviation value AL based on the sensor basic information S without using the sensor value detected by the additional sensor 411. val .

[0198] The information integrated by the sensor integration unit 420 is input to the correction unit 674. The correction unit 674 uses the correction rule acquired by the rule updating unit 671 and updates the deviation value AL determined by the basic value determining unit 673 based on the information provided from the sensor integration unit 420. val As a correction method, you can also directly correct the deviation value AL val In addition, the correction rule may be added to the rule used by the basic value determination unit 673 to determine the escape value AL. val .

[0199] In the sixth embodiment, even if the additional sensor unit 410 is added, the separation value AL is determined. val Therefore, when the additional sensor unit 410 is added, the decision deviation value AL is changed. val Compared with the situation of changing the entire rule, it is sufficient to make rule changes on a small scale.

[0200] <Seventh embodiment>

[0201] exist Fig.11 2 shows the structure of an automatic driving device 700 according to the seventh embodiment. The structure of a deviation value determination unit 772 included in a rule determination unit 770 of the automatic driving device 700 is different from that of the deviation value determination unit 672 of the sixth embodiment. The deviation value determination unit 772 includes a basic value determination unit 773 and a correction unit 674.

[0202] The basic value determination unit 773 determines the deviation value AL val The rule of the basic value determination unit 673 is the same as that of the sixth embodiment. However, the basic value determination unit 773 i Determine the escape value AL val After that, execute Fig.12 The processing shown is different from that of the basic value determination unit 673.

[0203] exist Fig.12 In S41, it is determined whether there are multiple escape values ​​AL val The candidate path T is 0 i If the result of S41 is "yes", the process proceeds to S42. In S42, the departure value AL determined by the candidate path is val The correction unit 674 inputs the deviation value AL corrected by the correction unit 674. When the process proceeds to S42, the deviation value AL corrected by the correction unit 674 is input to the correction unit 674. val The data is input to the controllability determination unit 173 .

[0204] If the judgment result of S42 is "No", that is, there is only one escape value AL val The candidate path T is the minimum value, that is, 0i In S43, the result determined by the basic value determination unit 773 is output to the control availability determination unit 173 instead of being input to the correction unit 674.

[0205] In the seventh embodiment, when there are a plurality of deviation values ​​AL determined by the basic value determination unit 773, val The candidate path T is 0 i In the case of (S41: Yes), the correction unit 674 corrects the deviation value AL determined by the basic value determination unit 773. val On the other hand, if there is only one deviation value AL determined by the basic value determination unit 773 val The candidate path T is 0 i (S41: No), the escape value AL determined by the basic value determination unit 773 is val The controllability determination unit 173 outputs the result (S43). This can reduce the calculation load compared to the case where the correction unit 674 always performs correction.

[0206] <Eighth Embodiment>

[0207] exist Fig.13 , the structure of the automatic driving device 800 of the eighth embodiment is shown. The structures of the sensor integration unit 850 and the rule determination unit 870 of the automatic driving device 800 are different from the sensor integration units 150, 420, and the rule determination units 170, 470, 570, 670, and 770 of the above-mentioned embodiments. The rule determination unit 870 of the eighth embodiment and the rule determination units 170, 470, 570, 670, and 770 of the above-mentioned embodiments are rule determination devices. When these rule determination devices execute processing, the rule determination method is executed.

[0208] The sensor integration unit 850 includes a rule management unit 855 in addition to the configuration of the sensor integration unit 150 of the first embodiment. The rule management unit 855 manages driving rules so that the driving rules corresponding to the driving conditions are applied as driving rules used for determination by the rule determination unit 870.

[0209] As described above, the driving condition includes the driving area. In addition, the driving condition includes various factors that may affect the driving of the vehicle 1. For example, the installation status of the sensor 101 is also included in the driving condition. When the sensor 101 is added, the range and accuracy of the surrounding area of ​​the vehicle 1 that can be detected change. Therefore, if the sensor 101 is added, it will affect the driving of the automatic driving. The failure and performance degradation of the sensor 101 are also included in the driving condition. This is because even if the sensor 101 fails or the performance is reduced, the range and accuracy of the surrounding area of ​​the vehicle 1 that can be detected will also change. Meteorology and weather can also be included in the driving condition. The specific route change condition described in detail later is also an example of a driving condition.

[0210] Traffic regulations may define an application start period, or an application start period and an application end period. When traffic regulations define an application period, whether or not the vehicle is within the application period of the traffic regulations is also an example of the driving condition.

[0211] The rule management unit 855 predicts the driving condition of the host vehicle 1. The rule management unit 855 obtains various information such as the position of the host vehicle 1, the direction of travel, the planned driving path of the host vehicle 1, and the conditions for applying the driving rules from the sensor 101, the server 2, etc., and predicts the driving condition of the host vehicle 1. The conditions for applying the driving rules include the driving area, the application period, etc.

[0212] The server 2 stores various driving rules, and functions as a rule distribution unit that distributes the stored driving rules. Since the driving rules have rules that define application conditions, the server 2 stores the driving rules in association with the application conditions of the driving rules.

[0213] In order to confirm whether there are driving rules with a relatively close application period, the rule management unit 855 may sequentially communicate with the server 2 via the wireless communication unit 140 to confirm whether there are newly applied driving rules. In addition, the rule management unit 855 may be notified of newly applied driving rules from the server 2 (i.e., push notification).

[0214] When the driving rule required in the predicted driving situation is not stored in the rule DB storage unit 130 , the rule management unit 855 causes the rule acquisition unit 154 to acquire the required driving rule in advance.

[0215] The rule acquisition unit 154 acquires the driving rule from the server 2 located outside the host vehicle 1 by wireless communication using the wireless communication unit 140 according to the instruction from the rule management unit 855. When the condition for applying the driving rule is stored in the server 2 in correspondence with the driving rule, the rule acquisition unit 154 also acquires the condition for applying the driving rule.

[0216] For example, when data specifying a region to which the driving rule is applied is stored in the server 2, the rule acquisition unit 154 also acquires the data specifying the region to which the driving rule is applied. When data specifying a period to which the driving rule is applied is stored in the server 2, the rule acquisition unit 154 also acquires the data specifying the period to which the driving rule is applied.

[0217] The driving rules acquired by the rule acquisition unit 154 may be driving rules of the same model as the driving rules already acquired but with different variables. For example, as an example of a variable, there is a speed limit in a model that is a concrete embodiment of the driving rules and a vehicle-to-vehicle distance specified by speed. In the case where the driving rules to be acquired are driving rules that differ from the already acquired driving rules only in variables, only the variables may be acquired. Of course, all driving rules including the variables may also be acquired.

[0218] The rule acquisition unit 154 stores the acquired driving rules and accompanying data in the rule DB storage unit 130. The rule management unit 855 sequentially determines the driving conditions, and when the driving rules used by the deviation value determination unit 872 need to be changed, provides the changed driving rules to the rule determination unit 870.

[0219] The rule determination unit 870 includes a controllability determination unit 173, a deviation value storage unit 174, an output unit 175, a rule change unit 871, a deviation value determination unit 872, and a path acquisition unit 876. The controllability determination unit 173, the deviation value storage unit 174, and the output unit 175 are the same as those described in the first embodiment. The path acquisition unit 876 acquires the candidate path T planned by the path planning unit 160. i In the above-described embodiments, the deviation value determination unit 171 , 472 , 672 , 772 has the function of the path acquisition unit 876 .

[0220] When the driving rule is provided from the rule management unit 855, the rule changing unit 871 changes the driving rule used by the deviation value determination unit 872 to the provided driving rule. In the above-mentioned embodiment, the minimum distance determination unit 172 and the deviation value determination units 171, 472, 672, and 772, which are the parts using the driving rule, have the functions of the rule management unit 855 and the rule changing unit 871.

[0221] The deviation value determination unit 872 determines the deviation value AL using the driving rule provided by the rule management unit 855. val . Departure value AL val The value indicating the result of judging whether the vehicle 1 complies with the driving rules. The rule management unit 855 provides the driving rules corresponding to the driving conditions to the rule judgment unit 870. The processing performed by the deviation value determination unit 872 is the same as that of the deviation value determination unit 472 except that the driving rules used are the driving rules provided by the rule management unit 855.

[0222] In the eighth embodiment, the driving rule used for determination by the departure value determination unit 872 is changed according to the driving condition of the host vehicle 1. Therefore, it is possible to determine whether the automatic driving control is appropriate in accordance with the change in the driving condition.

[0223] <Ninth embodiment>

[0224] exist Fig.14 The structure of the automatic driving device 900 of the ninth embodiment is shown in FIG. The structure of the sensor integration unit 950 of the automatic driving device 900 of the ninth embodiment is different from the sensor integration unit 850 of the eighth embodiment. In addition, for the convenience of illustration, Fig.14 The sensor 101 is omitted.

[0225] The sensor integration unit 950 includes a rule management unit 955 which is different from the rule management unit 855 of the eighth embodiment. Although not shown, the sensor integration unit 950 also includes a target vehicle behavior determination unit 151, a position acquisition unit 152, and an environment determination unit 153.

[0226] The rule management unit 955 includes a sensor state management unit 956 and a sensor monitoring range management unit 957. The sensor state management unit 956 manages the states of the sensor 101 and the additional sensor 411. Hereinafter, the states of the sensor 101 and the additional sensor 411 are referred to as sensor states.

[0227] The sensor status includes various factors that affect the signals detected by the sensor 101 and the additional sensor 411, such as the SN ratio, detection resolution, the presence of a fault, and the need for calibration. The sensor status management unit 956 sequentially determines whether there is a change in the sensor status that requires a change in the driving rules based on the signals provided from the sensor basic information S and the additional sensor unit 410.

[0228] An example of determining that there is a change in the sensor state that requires a change in the driving rules is described. For example, when a certain sensor 101 fails and a signal from the sensor 101 cannot be obtained, it can be determined that there is a change in the sensor state that requires a change in the driving rules. It can also be determined that there is a change in the sensor state that requires a change in the driving rules when the SN ratio of a certain sensor 101 is above a preset threshold and the detection resolution of a certain sensor 101 is below a preset threshold. It can also be determined that there is a change in the sensor state that requires a change in the driving rules when a preset recalibration period has exceeded after the sensor 101 has been calibrated.

[0229] The rule DB storage unit 130 may store driving rules that can be pre-assumed to correspond to the sensor state. When the sensor state management unit 956 determines that there is a change in the sensor state that requires a change in the driving rule, the rule management unit 955 causes the rule acquisition unit 154 to acquire a driving rule that is suitable for the changed sensor state. Furthermore, the rule management unit 955 provides the driving rule acquired by the rule acquisition unit 154 to the rule determination unit 870.

[0230] When the driving rule suitable for the changed sensor state is stored in the rule DB storage unit 130, the rule acquisition unit 154 acquires the driving rule suitable for the changed sensor state from the rule DB storage unit 130. The rule acquisition unit 154 may acquire the driving rule suitable for the changed sensor state from the server 2 via wireless communication.

[0231] The sensor monitoring range management unit 957 manages the monitoring range that can be monitored by the sensor 101 and the additional sensor 411. For example, when the additional sensor 411 for detecting the rear of the vehicle 1 is added, the sensor monitoring range management unit 957 adds the rear of the vehicle 1 to the monitoring range. When the sensor 101 for monitoring the front is equipped with a millimeter wave radar, but a LiDAR is added as the additional sensor 411 for monitoring the front, the sensor monitoring range management unit 957 updates the monitoring range of the front. This is because the monitoring ranges of the millimeter wave radar and the LiDAR are completely inconsistent.

[0232] The sensor monitoring range management unit 957 further determines in sequence whether the change in the surrounding range of the vehicle 1 that can be monitored by the sensor 101 and the additional sensor 411 has a change in the monitoring range that requires a change in the driving rules. In addition, the sensor monitoring range management unit 957 may be omitted, and the function of the sensor monitoring range management unit 957 may be implemented by the sensor state management unit 956.

[0233] The rule DB storage unit 130 can store driving rules that can be pre-assumed to correspond to the monitoring range. When the sensor monitoring range management unit 957 determines that there is a change in the monitoring range that requires a change in the driving rules, the rule management unit 955 causes the rule acquisition unit 154 to acquire driving rules that are suitable for the changed monitoring range. The rule management unit 955 provides the driving rules acquired by the rule acquisition unit 154 to the rule determination unit 870.

[0234] The rule acquisition unit 154 acquires the driving rule suitable for the changed monitoring range from the rule DB storage unit 130 or the server 2. An example of the driving rule suitable for the changed monitoring range is that when the changed monitoring range is a monitoring range in which the rear of the host vehicle 1 is added, the deviation value AL can be determined by including the behavior of an object existing behind the host vehicle 1. val Another example of a driving rule suitable for the changed monitoring range is, when the front can be monitored farther than before by adding LiDAR, the deviation value AL is determined by considering the behavior of surrounding vehicles in the range that can be monitored by LiDAR. val driving rules.

[0235] In the present embodiment, there is a case where the acquired driving rule is a driving rule of the same model as the already acquired driving rule but with different variables.

[0236] An example is described below. For example, a model that embodies the already acquired driving rules is a model that makes the most cautious judgment on a range that cannot be confirmed by the sensor 101. When the monitoring range is widened by adding the additional sensor 411, the acquired driving rules may have a variable that specifies the range for making the most cautious judgment, which corresponds to the widening of the monitoring range.

[0237] The driving rules acquired above are rules for making judgments based on sensor values ​​detected by multiple sensors respectively. When the sensor values ​​used for judgment cannot be obtained, the judgment on the sensor values ​​that cannot be obtained can also be said to be a rule for judging that the vehicle 1 does not follow the driving rules.

[0238] Furthermore, when the driving rules are changed to correspond to the case where the additional sensor 411 is added, the departure value determination unit 872 also determines the departure value AL using the sensor value detected by the additional sensor 411. val In other words, the rule determination unit 870 including the deviation value determination unit 872 also uses the sensor value detected by the additional sensor 411 to determine whether the host vehicle 1 deviates from the driving rule.

[0239] Furthermore, when a certain sensor 101 fails, contrary to the case where the additional sensor 411 is added, the acquired driving rule may be a driving rule in which a variable is changed in accordance with the monitoring range narrowed due to the failure of the sensor 101 .

[0240] In the ninth embodiment, whether there is a change in the sensor state or a change in the monitoring range that requires a change in the driving rules is managed. When there is a change in the sensor state or a change in the monitoring range that requires a change in the driving rules, a driving rule suitable for the changed sensor state or the changed monitoring range is acquired. Therefore, it is possible to determine whether it is appropriate automatic driving control corresponding to the change in driving conditions.

[0241] Although the embodiments have been described above, the disclosed technology is not limited to the above-described embodiments, and the following modifications are also included in the scope of disclosure. In addition, various modifications other than the following can be made and implemented without departing from the gist.

[0242] <Variation 1>

[0243] As in the sixth and seventh embodiments, the deviation value determination units 672 and 772 may be configured to include the basic value determination units 673 and 773 and the correction unit 674 , while a method may be applied in which a rule is provided for each driving area as in the first embodiment.

[0244] In this case, the basic value determination unit determines the deviation value AL according to a rule that does not depend on the driving area. val Then, the correction unit obtains a correction rule for each driving area, and adjusts the deviation value AL determined by the basic value determination unit according to the correction rule. val As a correction method, you can also directly correct the deviation value AL val In addition, the correction rule can be added to the rule used by the basic value determination unit to determine the escape value AL val .

[0245] In the embodiment of the first modification, when the travel area is changed, the departure value AL is changed. val Compared with the overall situation of the rules, rule changes become easier.

[0246] <Variation 2>

[0247] In the embodiment, a part of the driving rules is defined for each driving area as an example of a driving condition. However, a part of the driving rules may be defined by distinguishing between a specific lane change condition and a standard driving condition. In addition, the standard driving condition refers to a driving condition that does not conform to the specific lane change condition. In the case where a part of the driving rules is defined by distinguishing between a specific lane change condition and a standard driving condition, the above-described embodiment can be applied, similarly to the case where a part of the driving rules is defined for each driving area.

[0248] A specific lane change situation is a situation that satisfies the following two conditions. The first condition is that there is a section where the lane must be changed to the planned route. The second condition is that the probability that the vehicle distance is longer than the minimum distance allowed by the standard control mentioned above in the section where the lane change is necessary is lower than a preset threshold. A specific lane change situation refers to a situation where there is a high concern that the lane change cannot be made if the minimum distance allowed by the standard control is used. In addition, lane changes include lane changes. When driving in the same lane along the road, the lane change is not made even if the road is curved. Lane changes also include situations where the road on which the vehicle 1 is traveling changes. This is because if the road on which the vehicle 1 is traveling changes, the lane in which the vehicle 1 is traveling also changes. Therefore, merging is also included in lane changes.

[0249] It is also possible to directly determine whether the first condition and the second condition are satisfied and decide whether it is a specific route change situation. In addition, the second condition can also be said to be a determination of whether the road after the route change is congested in the section where the route change must be performed.

[0250] Therefore, it is also possible to determine whether the second condition is met by whether the road is congested after the route change in the section where the route change is necessary. The congestion here can be determined based on the length of the train that is traveling at a low speed. In addition, traveling at a low speed also includes repeated stops and starts. The low-speed traveling can be, for example, a speed of less than 40 km / h. Further, a lower speed, for example, less than 30 km / h or less than 20 km / h can also be considered as low-speed traveling. The length of the train is, for example, 1 km. In addition, a distance shorter than 1 km or a distance longer than 1 km can also be used as a congestion condition.

[0251] In addition, during congestion, the distance between vehicles is narrow, so the number of vehicles per unit road length increases. Therefore, congestion may be regarded as a case where the number of vehicles per unit road length is greater than a threshold value.

[0252] In this modification example 2, for example, it is possible to execute Fig.15 Processing shown. Fig.15The treatment shown is in place of Figure 7 The point that S34 is replaced by S34-1 and S37-1 is replaced by S37 is the same as Figure 7 S34-1 is executed when S33 is "Yes", that is, when it is determined that there is an area where lane change is not possible in automatic driving using the standard control minimum allowable distance. When S33 is "Yes", it can be said that the area is a specific lane change situation.

[0253] In S34-1, it is determined whether the control-allowed minimum distance for a specific route change condition allows automatic driving. Specifically, in the case of the traffic conditions obtained in S32, it is determined whether the possibility that the vehicle distance when the route is changed is longer than the control-allowed minimum distance determined based on the relationship for determining the control-allowed minimum distance under the specific route change condition is higher than the second threshold. The second threshold is set to a value that is the same as or greater than the first threshold. In addition, the relationship for determining the control-allowed minimum distance under a specific route change condition can be determined by learning in the same way as the determination rule for the control-allowed minimum distance for each driving area.

[0254] If the judgment result of S34-1 is "No", then enter Figure 5 S24. If the judgment result of S34-1 is "yes", enter S35. In S35, if the judgment result of S34-1 is "yes", the occupant of the vehicle 1 is asked whether the automatic driving control that controls the minimum allowed distance is allowed to be changed. Afterwards, when an instruction is given from the occupant to allow the automatic driving control that controls the minimum allowed distance to be changed (S36: yes), in S37-1, under a specific route change condition, the automatic driving is performed using the relationship of the minimum allowed distance for determining the specific route change condition. In addition, S35 and S36 are omitted, that is, it is also possible not to ask the occupant of the vehicle 1 whether the automatic driving control that controls the minimum allowed distance is allowed to be changed, and if the judgment result of S34-1 is "yes", S37-1 is executed.

[0255] <Variation 3>

[0256] In the fourth to seventh embodiments, the configuration in which the additional sensor 411 is added is described as one form of sensor modification. In addition to adding a sensor, sensor modification also includes sensor updating and sensor deletion.

[0257] <Variation 4>

[0258] Part of the functions of the automatic driving device described in the embodiment and the modified example may be located outside the host vehicle 1 (for example, the server 2 ).

[0259] <Variant 5>

[0260] The user may be notified of the change in driving regulations from a predetermined notification unit.

[0261] <Variant 6>

[0262] Departure value AL val It may indicate only one value of 0 or 1. One of 0 and 1 indicates that the host vehicle 1 complies with the driving regulations, and the other of 0 and 1 indicates that the host vehicle 1 does not comply with the driving regulations.

[0263] <Variant 7>

[0264] The sensor integration unit 150, the path planning unit 160, the rule judgment unit 170, 470, 570, 670, and the driving unit 180 are the control units mentioned below. The control unit and its methods described in the present disclosure can be implemented by a special-purpose computer, which constitutes a processor programmed to execute one or more functions concretized by a computer program. Alternatively, the control unit and its methods described in the present disclosure can also be implemented by a special-purpose hardware logic circuit. Alternatively, the control unit and its methods described in the present disclosure can also be implemented by one or more special-purpose computers, which are composed of a combination of a processor that executes a computer program and one or more hardware logic circuits. Hardware logic circuits are, for example, ASICs and FPGAs.

[0265] The storage medium storing the computer program is not limited to the ROM, and may be stored in a computer-readable non-transitory tangible recording medium as instructions executed by the computer. For example, the program may be stored in a flash memory.

Claims

1. An automatic driving device is an automatic driving device used in a vehicle capable of automatic driving, comprising: The deviation value determination unit determines, when the vehicle equipped with the automatic driving device, i.e., the host vehicle, is traveling on the candidate path planned by the path planning unit, a deviation value indicating the possibility that the host vehicle deviates from the driving rules for each of the candidate paths based on a comparison result between the vehicle-to-vehicle distance between the host vehicle and surrounding vehicles existing around the host vehicle and a minimum control allowable distance, wherein: The path planning unit sequentially plans the candidate paths, where the candidate paths are candidates for the paths that the host vehicle should travel next in order to continue traveling. a controllability determination unit, which selects a path whose deviation value determined by the deviation value determination unit according to each of the candidate paths is within a control permission range from among the candidate paths planned by the path planning unit as a path for the host vehicle to travel, and outputs an instruction to travel on the selected path to the driving unit, and the driving unit performs driving control for the host vehicle to travel; as well as The rule updating unit updates the rule used by the deviation value determining unit to determine the deviation value based on the changed sensor when a sensor detecting the behavior of at least one of the surrounding vehicles and the host vehicle is changed.

2. The automatic driving device according to claim 1, wherein: have: A reference escape value determination unit determines the escape value based on the rule before the change independently from the control possibility determination unit; and The output unit outputs the deviation value determined by the reference deviation value determination unit.

3. The automatic driving device according to claim 1, wherein: The above-mentioned escape value determination unit comprises: A basic value determination unit determines the deviation value according to a predetermined invariable rule; and a correction unit that corrects the deviation value determined by the basic value determination unit, The rule updating unit acquires a correction rule used by the correction unit.

4. The automatic driving device according to claim 3, wherein: The rule updating unit acquires the correction rule corresponding to the driving area. The correction unit corrects the deviation value determined by the basic value determination unit using the correction rule corresponding to the driving area.

5. The automatic driving device according to claim 3, wherein: When a sensor is added, the rule updating unit acquires a rule for correcting the deviation value based on the added sensor. The correction unit acquires a sensor value detected by the added sensor, and corrects the deviation value determined by the basic value determination unit based on the acquired sensor value.

6. The automatic driving device according to any one of claims 3 to 5, wherein: In the case where there are multiple paths with the lowest deviation value determined by the basic value determination unit, the deviation value determination unit corrects the deviation value determined by the basic value determination unit through the correction unit. On the other hand, if there is only one path with the lowest deviation value determined by the basic value determination unit, the deviation value determination unit outputs the deviation value determined by the basic value determination unit to the controllability judgment unit.

7. An automatic driving device, which is an automatic driving device used in a vehicle capable of automatic driving, comprising: The rule determination unit determines whether the vehicle equipped with the automatic driving device, that is, the vehicle itself, complies with the driving rules when the vehicle is traveling on the candidate path planned by the path planning unit, wherein: The path planning unit sequentially plans the candidate paths, where the candidate paths are candidates for the paths that the host vehicle should travel next in order to continue traveling. a rule management unit that manages the driving rule so that the driving rule corresponding to the driving condition is applied as the driving rule used for determination by the rule determination unit; and a rule changing unit that changes the driving rule used by the rule judging unit for judging according to the driving condition, The rule judgment unit performs the judgment based on sensor values ​​detected by a plurality of sensors. When the sensor values ​​used for the judgment cannot be obtained, the judgment on the sensor values ​​that cannot be obtained is performed on the basis that the vehicle does not follow the driving rules. When the sensor that detects the sensor values ​​that cannot be obtained is added, the judgment is performed using the sensor values ​​detected by the added sensors.

8. The automatic driving device according to claim 7, wherein: The rule management unit predicts the driving condition of the host vehicle. The automatic driving device includes a rule acquisition unit configured to acquire the driving rule corresponding to the driving situation predicted by the rule management unit.

9. The automatic driving device according to claim 8, wherein: The rule acquisition unit acquires the driving rule corresponding to the predicted driving condition from outside the host vehicle through wireless communication.

10. The automatic driving device according to claim 8, wherein: When the area to which the acquired driving rule is applied is specified, the rule acquisition unit also acquires data specifying the area to which the acquired driving rule is applied.

11. The automatic driving device according to claim 8, wherein: When the period during which the acquired driving rule is applied is determined, the rule acquisition unit also acquires data indicating the period during which the acquired driving rule is applied.

12. The automatic driving device according to claim 8, wherein: The rule acquisition unit acquires the driving rule having the same model as the already acquired driving rule but with different variables.

13. A rule judgment device, comprising: The rule determination unit determines whether the vehicle equipped with the automatic driving device, that is, the vehicle itself, complies with the driving rules when the vehicle is traveling on the candidate path planned by the path planning unit, wherein: The route planning unit sequentially plans the candidate routes, wherein the candidate routes are candidates for the routes that the host vehicle should travel next in order to continue traveling; and a rule changing unit that changes the driving rule used by the rule judging unit for judging according to the driving condition, The rule judgment unit performs the judgment based on sensor values ​​detected by a plurality of sensors. When the sensor values ​​used for the judgment cannot be obtained, the judgment on the sensor values ​​that cannot be obtained is performed on the basis that the vehicle does not follow the driving rules. When the sensor that detects the sensor values ​​that cannot be obtained is added, the judgment is performed using the sensor values ​​detected by the added sensors.

Images