Travel path estimation device
By integrating the external condition and driving state detection unit on the vehicle, estimating and determining the driving path to be driven, and storing it in association with map information, the problem of autonomous driving and driving assistance without pre-establishing the map information area is solved, and the automatic vehicle is improved Continuity and accuracy of driving and driving assistance.
Patent Information
- Application Number
- CN202280094408.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2022-03-30
- Publication Date
- 2024-11-12
AI Technical Summary
The prior art is difficult to realize autonomous driving and driving assistance of vehicles in areas without pre-established map information.
A driving road estimation device is designed, and the vehicle's surrounding conditions and driving state are detected by the external condition detection unit and the driving state detection unit. The controller stores map information and estimates the driving road that the vehicle should travel based on the detected external conditions and driving state. , and determine and store it in association with the actual trajectory to ensure the accuracy of the estimated route.
Even in areas without pre-established map information, the sustainability of the vehicle's autonomous driving and driving assistance can be improved, and the situation of inappropriate driving road estimates can be reduced, and the driver will be avoided.
Smart Images

Figure CN118946779A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a travel path estimating device for estimating a travel path that a vehicle having an automatic driving function and a driving assistance function should travel. Background Art
[0002] In the past, there is a known device for automatically driving a vehicle (see, for example, Patent Document 1). In the device described in Patent Document 1, three-dimensional point cloud data obtained during driving is compared with a pre-established three-dimensional point cloud map containing information such as roads, and three-dimensional point cloud data of static objects such as roads is obtained from three-dimensional image data generated by fusing two-dimensional image data obtained during driving and three-dimensional point cloud data.
[0003] The popularization of vehicles with autonomous driving and driver assistance functions can improve the safety and convenience of the entire transportation society and realize a sustainable transportation system. In addition, by improving the efficiency and smoothness of transportation, CO2 emissions can be reduced, thereby reducing the environmental load.
[0004] Prior art literature
[0005] Patent Literature
[0006] Patent document 1: Japanese Patent Application Publication No. 2020-85886. Summary of the invention
[0007] Problem that the invention aims to solve
[0008] It is desirable to be able to continue automatic driving and driving assistance of the vehicle even in an area without pre-established map information. However, when using pre-established map information as in the device described in Patent Document 1, it is difficult to continue automatic driving and driving assistance in such an area.
[0009] Solutions for solving problems
[0010] A driving path estimation device according to a technical solution of the present invention comprises: an external condition detection unit mounted on a vehicle to detect external conditions around the vehicle; a driving state detection unit to detect the driving state of the vehicle; and a controller. The controller stores map information, estimates the driving path that the vehicle should travel at a predetermined period based on the external conditions detected by the external condition detection unit, determines the track that the vehicle actually traveled based on the driving state detected by the driving state detection unit, determines whether the estimated driving path is consistent with the determined track, and stores the determination result in association with the map information.
[0011] Effects of the Invention
[0012] According to the present invention, even in an area without pre-established map information, the continuity of the vehicle's automatic driving and driving assistance can be improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] Figure 1 It is a block diagram schematically showing the configuration of a travel path estimating device according to an embodiment of the present invention.
[0014] Figure 2A This is a diagram showing an example of a traveling path when the estimated traveling path matches the actual trajectory.
[0015] Figure 2B This is a diagram showing an example of a trajectory when the estimated travel path matches the actual trajectory.
[0016] Figure 3A This is a diagram showing an example of a traveling path when the estimated traveling path does not match the actual trajectory.
[0017] Figure 3B This is a diagram showing an example of a trajectory when the estimated travel path does not coincide with the actual trajectory.
[0018] Figure 4 This is a diagram showing an example of a determination result stored in association with map information.
[0019] Figure 5 This is a diagram for explaining whether automatic driving or driving assistance can be performed based on the determination result.
[0020] Fig. 6A This is a diagram for explaining the exclusion of a travel path estimated at a specific time.
[0021] Figure 6B This is a diagram for explaining the generation of a driving road map.
[0022] Figure 7 It is a diagram for explaining the calculation of reliability based on the determination result. DETAILED DESCRIPTION
[0023] Below, refer to Figures 1 to 7 An embodiment of the present invention is described. The driving path estimation device of the embodiment of the present invention is applicable to a vehicle having a driving assistance function of controlling a driving actuator to assist the driver of the vehicle or to enable the vehicle to drive automatically, and estimates the driving path that the vehicle should travel. "Driving assistance" in this embodiment includes driving assistance that assists the driver's driving operation and automatic driving that enables the vehicle to drive automatically without the help of the driver's driving operation, which is equivalent to Level 1 to Level 4 automatic driving defined by SAE, and "automatic driving" is equivalent to Level 5 automatic driving.
[0024] In autonomous driving and driving assistance, since the steering assistance of assisted steering has a great effect in reducing the burden on the driver, it is required to continue to provide assistance not only on highways with well-marked roads that specify driving lanes, but also at intersections on general roads with incomplete road markings. However, although high-precision maps for autonomous driving and driving assistance are produced in areas with high traffic volume such as highways and cities, they are not produced in areas with low traffic volume such as residential areas and suburbs. In addition, after the latest high-precision map is produced, the road structure may sometimes change due to construction, etc.
[0025] Therefore, even in areas where a high-precision map has not been created in advance, in order to improve the continuity of automatic driving and driving assistance, it is necessary to use the vehicle-mounted sensors and the like on the vehicle side to estimate the driving path to be traveled when each vehicle is traveling. Therefore, in this embodiment, the driving path estimation device is configured as follows, so that the driving path to be traveled by the vehicle is estimated by using the vehicle-mounted sensors and the like, thereby improving the continuity of automatic driving and driving assistance of the vehicle even in areas where there is no pre-established map information.
[0026] Figure 1 2 is a block diagram schematically showing the structure of the travel path estimation device 10 according to the embodiment of the present invention. Figure 1 As shown, the travel path estimation device 10 includes an external condition detection unit 2, a travel state detection unit 3, a travel actuator 4, and a controller 5 mounted on the vehicle 1. The controller 5 is connected to the external condition detection unit 2, the travel state detection unit 3, and the travel actuator 4.
[0027] The external condition detection unit 2 is mounted on the vehicle 1 to detect the external conditions around the vehicle 1, especially in front. The external condition detection unit 2 includes, for example, a camera having a CCD (charge coupled device), a CMOS (complementary metal oxide semiconductor) and other shooting elements to shoot the surroundings of the vehicle 1. The external condition detection unit 2 may also include a millimeter wave radar, which irradiates millimeter waves (radio waves) and measures the distance and direction to the object based on the time it takes for the irradiated waves to hit the object and return. The external condition detection unit 2 may also include a laser radar (LiDAR), which irradiates lasers and measures the distance and direction to the object based on the time it takes for the irradiated light to hit the object and return.
[0028] Based on the external conditions detected by the external condition detection unit 2, the road surface on which the vehicle can travel, the driving lane defined by road markings and structures, obstacles on the road including traffic participants such as surrounding vehicles and pedestrians, etc. can be identified, and the driving path EL that the vehicle 1 should travel can be estimated. More specifically, the driving path EL(t) from the current position P(t) of the vehicle 1 at a predetermined time t to at least a predetermined distance ahead that can be detected by the external condition detection unit 2 can be estimated.
[0029] The driving state detection unit 3 detects the driving state of the vehicle 1, such as the driving speed and the traveling direction. The driving state detection unit 3 includes, for example, an inertial measurement unit (IMU) that detects the rotational angular velocity of three axes, namely, the vertical direction around the center of gravity of the vehicle 1, the traveling direction, and the vehicle width direction, and the acceleration in the three-axis directions. The driving state detection unit 3 may also include a wheel speed sensor that detects the rotational speed of each wheel of the vehicle 1. The driving state detection unit 3 may also include a positioning unit that determines the current position (latitude, longitude) of the vehicle 1 based on a positioning signal from a positioning satellite, etc.
[0030] Based on the driving state detected by the driving state detection unit 3, the movement amount of the vehicle 1 at each predetermined time can be calculated, thereby determining the trajectory AL traveled by the vehicle 1. For example, the actual trajectory AL(t) traveled by the vehicle 1 from the predetermined time t when the driving path EL of the vehicle 1 is estimated to the current time can be determined. The actual trajectory AL traveled by the vehicle 1 can also be determined by connecting the vehicle positions at each predetermined time.
[0031] The travel actuator 4 includes a steering mechanism such as a steering gear for steering the vehicle 1 , a driving mechanism such as an engine and a motor for driving the vehicle 1 , and a braking mechanism such as a brake for braking the vehicle 1 .
[0032] The controller 5 is composed of a computer including a processing unit 6 such as a CPU (central processing unit), a storage unit 7 such as a RAM (random access memory), a ROM (read only memory), an I / O interface, and other peripheral circuits. The controller 5 is, for example, a part of a group of multiple electronic control units (ECUs) mounted on the vehicle 1 and controlling the operation of the vehicle 1. In the storage unit 7 of the controller 5, general map information for the navigation device to guide the vehicle 1 from the current location to the destination is pre-stored. The storage unit 7 also stores high-precision maps for automatic driving.
[0033] Figure 2A and Figure 2B 1 is a diagram showing an example in which the travel path EL(t) estimated at a predetermined time t coincides with the actual trajectory AL(t). Figure 3A and Figure 3B This is a diagram showing an example in which the travel path EL(t) and the trajectory AL(t) do not coincide with each other.
[0034] like Figure 2A and Figure 3AAs shown, the processing unit 6 of the controller 5 estimates the travel path EL that the vehicle 1 should travel based on the external conditions detected by the external condition detection unit 2. More specifically, the travel path EL(t) from the current position P(t) of the vehicle 1 at a predetermined time t to the predetermined distance ahead is estimated. Such a travel path EL is estimated at a predetermined period (for example, the detection period of the external condition detection unit 2, the control period of the controller 5, the communication period between the external condition detection unit 2 and the controller 5, etc.).
[0035] The information of the estimated travel path EL(t) is stored in association with the map information in the storage unit 7. More specifically, the information is associated with the current position P(t) of the vehicle 1 at the time t when the travel path EL(t) is estimated and added to the map information.
[0036] like Figure 2B and Figure 3B As shown, the processing unit 6 of the controller 5 determines the actual trajectory AL when the vehicle 1 is driven with the participation of the driver based on the driving state detected by the driving state detection unit 3, and determines whether the determined trajectory AL is consistent with the estimated driving path EL. More specifically, the actual trajectory AL(t) when the vehicle 1 is driven with the participation of the driver in the section from the current position P(t) of the vehicle 1 at the time t when the driving path EL(t) is estimated to the predetermined distance is determined, and it is determined whether it falls within the range of the estimated driving path EL(t).
[0037] It should be noted that the driving in which the driver participates includes driving when the driver does not accept driving assistance and performs manual driving and driving when the driver accepts driving assistance and performs manual driving.
[0038] like Figure 2B As shown, when the actual trajectory AL(t) of the vehicle 1 while driving with the driver involved falls within the range of the estimated driving path EL(t), it is determined that the estimated driving path EL(t) is consistent with the determined trajectory AL(t). Alternatively, it is determined that the driving path estimation is appropriate.
[0039] On the other hand, Figure 3B As shown in FIG. 1 , when the actual trajectory AL(t) of the vehicle 1 when the driver is involved in driving does not fall within the range of the estimated driving path EL(t), it is determined that the estimated driving path EL(t) is inconsistent with the determined trajectory AL(t). In this way, when the estimated driving path EL is inconsistent with the trajectory AL when the driver is involved in driving, the possibility of inappropriate driving path estimation is high, and therefore it is treated as a scene in which the external condition detection unit 2 is not good at estimating the driving path.
[0040] Such a determination result is associated with the map information together with the information of the estimated travel path EL(t) and stored in the storage unit 7. More specifically, it is associated with the position information of the current position P(t) of the vehicle 1 at the time t when the travel path EL(t) is estimated, and is added to the map information together with the information of the estimated travel path EL(t).
[0041] When the estimated travel route EL is inconsistent with the trajectory AL when the driver is actually traveling, the detection results of the external condition detection unit 2 at the time t or before and after the estimated travel route EL(t) may be sent to an external analysis device, etc. In this case, only data of scenes where the external condition detection unit 2 is not good at estimating the travel route can be effectively collected from multiple vehicles 1, and the main reason for inappropriate travel route estimation can be analyzed.
[0042] Figure 4 FIG. 4 is a diagram showing an example of a determination result stored in association with map information. Figure 4 As shown, in the map information stored in the storage unit 7 of the controller 5, corresponding to the vehicle position of each specified period in the area where the vehicle 1 actually traveled with the driver, information on the judgment result of whether the scene is one in which the external condition detection unit 2 is not good at estimating the travel path is attached.
[0043] Figure 5 This diagram is used to explain whether automatic driving or driving assistance can be performed based on a determination result, and shows a situation in which the vehicle 1 uses automatic driving or driving assistance to travel in an area where the vehicle 1 has traveled in the past with the participation of a driver.
[0044] like Figure 5 As shown, the processing unit 6 of the controller 5 estimates the travel path EL based on the external conditions detected by the external condition detection unit 2, and determines whether to continue or interrupt the automatic driving and driving assistance in use based on the past determination results of the vehicle 1 on the road ahead. In other words, it is determined whether to allow the control of the travel actuator 4 based on the travel path EL estimated based on the external conditions detected by the external condition detection unit 2 at each point on the road ahead.
[0045] More specifically, it is determined whether there is a place that was estimated to have been on the travel path EL in the past and determined to be an unsuitable scene in a predetermined section on the road ahead of the vehicle 1 (for example, from the current position of the vehicle 1 to the section about 300 meters ahead). If there is a place determined to be an unsuitable scene in the predetermined section, the currently used automatic driving or driving assistance is interrupted after notifying the driver. The interruption of the currently used automatic driving or driving assistance also includes downgrading from automatic driving to driving assistance and downgrading of driving assistance.
[0046] For example, when steering assistance is performed based on an inappropriate driving path EL, an auxiliary steering torque in an inappropriate direction is applied to the steering wheel held by the driver, and the driver needs to correct it, which becomes a burden on the driver. When approaching a location that is estimated to be a high possibility of an inappropriate driving path EL, the driver is notified in front of the location, and then the automatic driving and driving assistance are interrupted, thereby preventing unnecessary burden on the driver. It should be noted that the length of the prescribed interval for determining whether automatic driving and driving assistance can be performed can also be changed according to the road shape, speed limit, etc. on the road ahead.
[0047] The detection of external conditions by the external condition detection unit 2 is limited by factors such as the surrounding environment, so it is difficult to always infer the appropriate driving route EL at some locations. By pre-registering such locations as poor scenarios, it is possible to actively use the driving route estimation of the external condition detection unit 2 in addition to poor scenarios. For example, when using DNN (Deep Neural Network) to identify the driving route EL based on the detection results of the external condition detection unit 2, when a strict threshold is set to only identify the driving route EL, some locations cannot identify the driving route EL itself and cannot use the driving route estimation. By setting a loose threshold that allows some locations to also identify non-driving routes EL, and pre-registering locations that are more likely to identify inappropriate routes as poor scenarios, the availability of the route estimation of the entire device can be improved.
[0048] Fig. 6A is a diagram for explaining the elimination of the travel path EL estimated at specific times t1 to t4. Figure 6B MAP is a diagram for explaining the generation of the driving road map MAP. Fig. 6A and Figure 6B As shown, the processing unit 6 of the controller 5 determines whether to allow the generation of the travel path map MAP based on the estimated travel path EL based on the determination result stored in the storage unit 7 in association with the map information.
[0049] More specifically, for each location where the vehicle 1 has traveled in the past with the participation of the driver, it is determined whether it is an unfamiliar scene, and the travel route EL estimated outside the unfamiliar scene is allowed to be used to generate the travel route map MAP, while the travel route EL estimated in the unfamiliar scene is prohibited. Fig. 6A and Figure 6B In the example, the driving route EL(t3) estimated at the time t3 at the location determined as the unfavorable scene is excluded, and the driving route map MAP is generated based on the driving routes EL(t1), EL(t2), and EL(t4) estimated at the other locations at the times t1, t2, and t4. The generated driving route map MAP is associated with the map information and stored in the storage unit 7.
[0050] In the automatic driving of the vehicle 1, the consistency between the shape of the latest driving path EL estimated at any time during driving and the shape of the driving path EL stored as the driving path map MAP (or high-precision map) in the storage unit 7 is confirmed. The processing unit 6 of the controller 5 controls the driving actuator 4 to make the vehicle 1 travel within the estimated driving path EL based on the consistency between the shape of the latest driving path EL and the driving path EL on the driving path map MAP (or high-precision map).
[0051] Figure 7 is a diagram for explaining the calculation of the reliability based on the determination result, showing the case where a vehicle has traveled multiple times in the same place ( Figure 7 An example of the reliability calculated after 4 times. Figure 7 As shown, the processing unit 6 of the controller 5 calculates the reliability of the estimated travel path EL based on the determination result stored in association with the map information, and stores the calculated reliability in association with the map information.
[0052] The reliability can be calculated, for example, as the ratio of the actual number of times the appropriate driving route EL is estimated to the number of times the driving route EL is in the past. Figure 7 In the example, the calculation is performed such that the reliability of estimating a location on the appropriate driving path EL 4 times out of the past 4 times is 100%, the reliability of estimating a location on the appropriate driving path EL 3 times is 75%, and the reliability of not estimating a location on the appropriate driving path EL once is 0%.
[0053] In particular, when autonomous driving is performed, it is necessary to perform a highly reliable route estimation. The reliability of the route estimation is calculated based on actual situations when the vehicle is driven multiple times without autonomous driving but with the driver involved. The route estimation result is used only when the reliability exceeds an appropriate threshold, thereby enabling appropriate autonomous driving based on the highly reliable route estimation.
[0054] In addition, even if a scene was determined to be poor in the past due to some circumstances, as long as the number of times the estimated driving route is determined to be appropriate during multiple driving, it can be used as the estimated driving route EL in the driving scene with high reliability. In this way, the missing (interrupted) parts of the area where the estimated driving route can be used can be reduced, and the continuity of automatic driving and driving assistance can be improved.
[0055] According to this embodiment, the following effects can be achieved.
[0056] (1) The travel path estimation device 10 comprises: an external condition detection unit 2, which is mounted on the vehicle 1 and detects the external condition around the vehicle 1; a travel state detection unit 3, which detects the travel state of the vehicle 1; and a controller 5 ( Figure 1 The controller 5 stores the map information, estimates the travel path EL that the vehicle 1 should travel on a predetermined cycle based on the external conditions detected by the external condition detection unit 2, determines the trajectory AL that the vehicle 1 actually travels on based on the driving state detected by the driving state detection unit 3, determines whether the estimated travel path EL is consistent with the determined trajectory AL, and stores the determination result in association with the map information ( Figure 4 ).
[0057] That is, when the estimated driving route EL is inconsistent with the trajectory AL when the driver is actually driving, it can be identifiably registered as a poor scenario where there is a high possibility that the driving route estimation by the external condition detection unit 2 is inappropriate. As a result, the driving route EL estimated outside the poor scenario can be actively used, and the continuity of the automatic driving and driving assistance of the vehicle 1 can be improved even in an area without pre-established map information.
[0058] (2) The travel path estimation device 10 further includes a travel actuator 4 ( Figure 1 The controller 5 determines whether to allow the driving actuator 4 to be controlled according to the estimated driving path EL on the road ahead of the vehicle 1, based on the determination result stored in association with the map information. Figure 5 ). That is, the estimated driving path EL outside the unfamiliar scene can be actively used for automatic driving and driving assistance, which can improve sustainability. In addition, when there is an unfamiliar scene on the road ahead, the driver can be notified in advance and the automatic driving and driving assistance are interrupted, which can avoid inappropriate steering assistance and the burden on the driver related to its correction.
[0059] (3) The controller 5 determines whether to allow the generation of the travel path map MAP based on the estimated travel path EL, based on the determination result stored in association with the map information. Fig. 6A , Figure 6B That is, by excluding the estimated driving path EL in the unfavorable scene, a correct driving path map MAP can be generated. In addition, by making the driving path EL estimated in places other than the unfavorable scene actively used to generate the driving path map MAP, the result is that the continuity of automatic driving and driving assistance can be improved.
[0060] (4) The controller 5 calculates the reliability of the estimated travel path EL based on the determination result stored in association with the map information, and stores the calculated reliability in association with the map information. Figure 7). That is, the reliability of the estimated driving path in the driving scene is calculated based on multiple determination results when driving in the same driving scene. Thus, for example, only the driving path EL estimated in the driving scene with high reliability can be used, which is determined to be appropriate many times. In addition, even if it is determined to be an unfavorable scene due to special circumstances, as long as the number of determinations that the estimated driving path is appropriate is large, it can be used as the driving path EL estimated in the driving scene with high reliability, thereby further improving the continuity of automatic driving and driving assistance.
[0061] The above description is merely an example, and the above embodiment and modified examples do not limit the present invention unless the features of the present invention are impaired. The above embodiment and modified examples may be arbitrarily combined with one or more of them, and the modified examples may be combined with each other.
[0062] Description of Reference Numerals
[0063] 1: Vehicle;
[0064] 2: External condition detection department;
[0065] 3: Driving status detection unit;
[0066] 4: Actuator for driving;
[0067] 5: Controller;
[0068] 6: Processing department;
[0069] 7: Storage unit;
[0070] 10: Driving path estimation device;
[0071] AL: trajectory;
[0072] EL: driving road;
[0073] P: Current location.
Claims
1. A driving path estimation device, characterized in that: have: An external condition detection unit, which is mounted on the vehicle and detects external conditions around the vehicle; a driving state detecting unit for detecting the driving state of the vehicle; as well as Controller, The controller stores map information; estimates the driving path that the vehicle should travel at a predetermined period based on the external conditions detected by the external condition detection unit; and determines the actual driving track of the vehicle based on the driving state detected by the driving state detection unit; determining whether the estimated travel path is consistent with the determined trajectory; The determination result is stored in association with the map information.
2. The driving path estimation device according to claim 1, characterized in that: Also provided is a driving actuator mounted on the vehicle, The controller determines whether to permit control of the travel actuator based on the travel path estimated on the forward path of the vehicle, based on the determination result stored in association with the map information.
3. The driving path estimation device according to claim 1 or 2, characterized in that: The controller determines whether to allow generation of a driving road map based on the estimated driving road, based on the determination result stored in association with the map information.
4. The driving path estimating device according to any one of claims 1 to 3, characterized in that: The controller calculates the reliability of the estimated travel path based on the determination result stored in association with the map information, and stores the calculated reliability in association with the map information.
