Unmanned vehicle trajectory planning method, device, equipment and storage medium

Abstract

The present invention discloses a method, device, equipment and storage medium for unmanned vehicle trajectory planning. The method obtains the current vehicle state information, environmental perception information, global map information and trajectory planning result information of the current unmanned vehicle at the previous moment, performs risk assessment on the surrounding traffic environment of the current vehicle, and obtains the risk assessment result; calculates the time interval threshold of re-planning according to the risk assessment result, judges the event state in a preset planning trigger event set according to the time interval threshold, and obtains the event trigger judgment result; selects one of the modes of executing trajectory complete re-planning, speed re-planning or not executing re-planning according to the event trigger judgment result, and can ensure the stability of trajectory planning and high calculation efficiency, while having an adaptability similar to that of a planning method using a time trigger mechanism in a dynamic and complex environment, thereby improving the stability of trajectory planning and the comfort of riding.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned driving path planning, and in particular to an unmanned driving vehicle trajectory planning method, device, equipment and storage medium. Background Art

[0002] As a cutting-edge technology for today's vehicles, driverless technology has received widespread attention in recent years; motion planning, as a key core technology of driverless vehicles, directly determines the motion performance of driverless vehicles; during the actual driving process of driverless vehicles, since the external environment is constantly changing dynamically, the motion planning module of the driverless vehicle needs to be responsible for timely planning a drivable trajectory from the current position of the vehicle to the target position.

[0003] At present, the mainstream unmanned vehicle trajectory planning method is the time trigger mechanism, that is, the unmanned vehicle motion planning module plans a traversable trajectory from the current position to the target position in real time at a fixed period (generally 100ms) based on the received real-time environmental perception information and the current position status information of the vehicle. The time trigger mechanism can obtain high-frequency updated trajectory planning results. Since each trajectory planning is based on the real-time environmental perception and status information of the vehicle, it can significantly improve the adaptability of the unmanned vehicle in dynamic and complex environments. However, high-frequency triggered planning may lead to inconsistent behavior of trajectory planning results at adjacent moments, unstable motion of the unmanned vehicle (for example, the trajectory trend of the planned trajectory at the previous moment is to the left, and the trajectory trend of the trajectory at the next moment is to the right, and this inconsistent behavior phenomenon occurs frequently), and problems such as difficulty in tracking control. In addition, when the environment around the vehicle and the status of the vehicle have not changed much, the trajectory results planned within a certain period of time are often not much different. At this time, the high-frequency update of the trajectory will generate a large number of unnecessary operations, which seriously affects the calculation efficiency of the algorithm and increases the calculation burden of the platform.

[0004] One of the existing solutions is to determine whether at least one preset local path replanning event is met based on the traffic status information and driving status information of the unmanned vehicle: if it is determined that at least one local path replanning event is met, the local path of the vehicle is replanned; otherwise, no replanning is performed, and the vehicle continues to travel along the previously planned local path; however, the preset replanning trigger events of the above patent are incomplete and fail to fully consider the needs of various driving conditions, so it is difficult to cope with actual driving scenarios. At the same time, due to the overly conservative event settings and triggers, the unmanned vehicle cannot replan in time for potential risks in a dynamic and complex environment, which is prone to planning failures and even traffic accidents; and, when replanning, the patent only considers path replanning at the spatial level, but does not consider speed replanning at the temporal level, which further causes the unmanned vehicle to be overly conservative in the face of dynamic obstacles, low traffic efficiency, and difficult to cope with real complex road environments.

[0005] Another solution is to target lane-changing scenarios by using whether a trajectory collision occurs as a planning trigger event. If a trajectory collision occurs, speed re-planning and path re-planning are performed sequentially to ensure the lane-changing safety and traffic efficiency of the unmanned vehicle. However, this patent only targets lane-changing scenarios for unmanned vehicles and cannot handle cruising, obstacle avoidance and other scenarios for unmanned vehicles. At the same time, the preset planning trigger events are conservative (this patent only uses whether a trajectory collision occurs as a trigger event), and cannot take into account both the safety and real-time nature of trajectory planning. Summary of the invention

[0006] The main purpose of the present invention is to provide a method, device, equipment and storage medium for unmanned vehicle trajectory planning, aiming to solve the technical problems in the prior art that unmanned vehicles are too conservative in behavior when facing dynamic obstacles, have low traffic efficiency, are difficult to cope with real complex road environments, and cannot take into account the safety and real-time performance of trajectory planning.

[0007] In a first aspect, the present invention provides a method for unmanned vehicle trajectory planning, the method comprising the following steps:

[0008] Obtaining the current vehicle state information, environmental perception information, global map information, and last moment trajectory planning result information of the unmanned current vehicle, and performing a risk assessment on the surrounding traffic environment of the current vehicle based on the vehicle state information, the environmental perception information, the global map information, and the last moment trajectory planning result information to obtain a risk assessment result;

[0009] Calculate the time interval threshold of re-planning according to the risk assessment result, judge the event status in the preset planning trigger event set according to the time interval threshold, and obtain the event trigger judgment result;

[0010] A mode of executing complete trajectory replanning, speed replanning or not executing replanning is selected according to the event triggering judgment result.

[0011] Optionally, the obtaining of the current vehicle state information, environmental perception information, global map information, and trajectory planning result information of the unmanned current vehicle at a previous moment, and performing a risk assessment on the surrounding traffic environment of the current vehicle according to the vehicle state information, the environmental perception information, the global map information, and the trajectory planning result information of the previous moment to obtain a risk assessment result includes:

[0012] Obtain the current vehicle status information, environmental perception information, global map information and trajectory planning result information of the unmanned vehicle at the previous moment;

[0013] Determine the number of obstacles in a preset vehicle body area of ​​the current vehicle at a current vehicle speed, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles according to the vehicle state information, the environment perception information, the global map information, and the trajectory planning result information at the previous moment;

[0014] The risk assessment of the surrounding traffic environment of the current vehicle is performed according to the number of obstacles, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles to obtain a risk assessment result.

[0015] Optionally, performing risk assessment on the surrounding traffic environment of the current vehicle according to the number of obstacles, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles to obtain a risk assessment result includes:

[0016] According to the number of obstacles, the obstacle number risk sub-function is determined by the following formula:

[0017]

[0018] in, is the risk subfunction of the number of obstacles, is the number of obstacles within the current vehicle at the current moment, is the normalization coefficient;

[0019] According to the relative distance between the obstacle and the vehicle, the relative distance risk subfunction is determined by the following formula:

[0020]

[0021] in, is the relative distance risk subfunction, is the normalization coefficient, For the The lateral distance between the obstacle and the current vehicle, For the The longitudinal distance between the obstacle and the current vehicle, is the lateral distance weight coefficient, is the longitudinal distance weight coefficient;

[0022] According to the obstacle motion state, the motion state risk subfunction is determined by the following formula:

[0023]

[0024] in, is the motion state risk subfunction, is the normalization coefficient, is the obstacle motion state;

[0025] A traffic environment risk function is determined according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function, and a risk assessment is performed on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result.

[0026] Optionally, determining a traffic environment risk function according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function, and performing risk assessment on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result includes:

[0027] The traffic environment risk function is determined by the following formula according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function:

[0028]

[0029] in, is the traffic environment risk function, is the absolute time moment, is the obstacle quantity risk subfunction The weight coefficient of is the relative distance risk subfunction The weight coefficient of is the motion state risk subfunction The weight coefficient of

[0030] A risk assessment is performed on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result.

[0031] Optionally, the step of calculating a time interval threshold for re-planning according to the risk assessment result, judging an event state in a preset planning triggering event set according to the time interval threshold, and obtaining an event triggering judgment result includes:

[0032] The re-planning time interval threshold is calculated according to the risk assessment result by the following formula:

[0033]

[0034] in, is the re-planning time interval threshold, For preview duration, is the traffic environment risk function;

[0035] Obtain the start time corresponding to the start of the current plan and the time when the most recent trajectory replanning occurred from the preset plan trigger event set;

[0036] Compare the time difference between the start time and the occurrence time with the duration interval threshold, and when the time difference is greater than the duration interval threshold, determine that the event state is an event in which the time interval between the current moment and the last replanning is greater than the duration interval threshold;

[0037] Generate a first planning trigger event set according to the first planning start event, the time interval between the current moment and the last re-planning is greater than the time interval threshold event, the vehicle driving state switching event, and the distance deviation between the current position of the vehicle and the nearest point on the trajectory at the last moment is greater than the preset distance threshold event;

[0038] Taking the trajectory collision detection failure event as the second planning trigger event set;

[0039] An event status judgment is performed on the first planning triggering event set and the second planning triggering event set to obtain an event triggering judgment result.

[0040] Optionally, performing event status judgment on the first planning triggering event set and the second planning triggering event set to obtain an event triggering judgment result includes:

[0041] When at least one event in the first planning trigger event set occurs, the first trigger event is assigned a value of true, and the state judgment of the second planning trigger event set is terminated;

[0042] If none of the events in the first planned trigger event set occurs, the first trigger event is assigned a value of false, and the judgment of the second planned trigger event set is executed sequentially;

[0043] When at least one event in the second planned trigger event set occurs, assigning the second trigger event a value of true;

[0044] If none of the events in the second planned trigger event set occurs, the second trigger event is assigned a value of false;

[0045] Complete event trigger judgment and generate event trigger judgment results.

[0046] Optionally, the selecting, according to the event triggering judgment result, to execute one of the modes of trajectory complete replanning, speed replanning, or not executing replanning includes:

[0047] When the event triggering judgment result is that both the first triggering event and the second triggering event are false, re-planning is not performed;

[0048] When the event triggering judgment result is that the first triggering event is true, performing trajectory complete replanning;

[0049] When the second trigger events are both true, speed replanning is performed.

[0050] In a second aspect, in order to achieve the above-mentioned object, the present invention further proposes a trajectory planning device for an unmanned vehicle, the trajectory planning device for an unmanned vehicle comprising:

[0051] A risk assessment module is used to obtain the current vehicle state information, environmental perception information, global map information and trajectory planning result information of the unmanned current vehicle, and to perform a risk assessment on the surrounding traffic environment of the current vehicle based on the vehicle state information, the environmental perception information, the global map information and the trajectory planning result information of the previous moment to obtain a risk assessment result;

[0052] An event trigger judgment module is used to calculate a time interval threshold for re-planning according to the risk assessment result, and to judge the event status in the preset planning trigger event set according to the time interval threshold to obtain an event trigger judgment result;

[0053] The planning execution module is used to select a mode of executing trajectory complete replanning, speed replanning or not executing replanning according to the event trigger judgment result.

[0054] In the third aspect, in order to achieve the above-mentioned purpose, the present invention also proposes an unmanned vehicle trajectory planning device, which includes: a memory, a processor, and an unmanned vehicle trajectory planning program stored in the memory and executable on the processor, and the unmanned vehicle trajectory planning program is configured to implement the steps of the unmanned vehicle trajectory planning method described above.

[0055] In a fourth aspect, in order to achieve the above-mentioned purpose, the present invention also proposes a storage medium, on which an unmanned vehicle trajectory planning program is stored, and when the unmanned vehicle trajectory planning program is executed by a processor, the steps of the unmanned vehicle trajectory planning method described above are implemented.

[0056] The unmanned vehicle trajectory planning method proposed by the present invention obtains the current vehicle state information, environmental perception information, global map information and trajectory planning result information of the unmanned current vehicle at the current moment, and performs risk assessment on the surrounding traffic environment of the current vehicle according to the vehicle state information, the environmental perception information, the global map information and the trajectory planning result information at the previous moment to obtain the risk assessment result; calculates the time interval threshold of re-planning according to the risk assessment result, and judges the event state in the preset planning trigger event set according to the time interval threshold to obtain the event trigger judgment result; selects one of the modes of executing trajectory complete re-planning, speed re-planning or not executing re-planning according to the event trigger judgment result, while ensuring the stability of trajectory planning and high calculation efficiency, it has an adaptability similar to that of the planning method using the time trigger mechanism in a dynamic and complex environment; and by constructing a complete unmanned vehicle driving state abnormal event set as the trigger condition for planning, it can cope with and handle various automatic driving scenarios, can effectively reduce the number of unnecessary re-planning times, improve algorithm efficiency, reduce the calculation burden of the computing platform, and improve the stability of trajectory planning and riding comfort. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention;

[0058] Figure 2 A schematic diagram of a flow chart of a first embodiment of a method for trajectory planning of an unmanned vehicle according to the present invention;

[0059] Figure 3 A schematic diagram of vehicle position distance deviation in the unmanned vehicle trajectory planning method of the present invention;

[0060] Figure 4 A schematic diagram of a trajectory collision detection process in the unmanned vehicle trajectory planning method of the present invention;

[0061] Figure 5 A schematic diagram of event state judgment in the unmanned vehicle trajectory planning method of the present invention;

[0062] Figure 6 A schematic diagram of a flow chart of a second embodiment of a method for trajectory planning of an unmanned vehicle according to the present invention;

[0063] Figure 7 A schematic diagram of real-time traffic environment risk modeling in the unmanned vehicle trajectory planning method of the present invention;

[0064] Figure 8 This is a functional module diagram of the first embodiment of the unmanned vehicle trajectory planning device of the present invention.

[0065] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0066] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0067] The solution of the embodiment of the present invention is mainly: by obtaining the current vehicle state information, environmental perception information, global map information and trajectory planning result information of the unmanned current vehicle at the current moment, the risk assessment of the surrounding traffic environment of the current vehicle is performed according to the vehicle state information, the environmental perception information, the global map information and the trajectory planning result information of the previous moment, and the risk assessment result is obtained; the time interval threshold of re-planning is calculated according to the risk assessment result, and the event state in the preset planning trigger event set is judged according to the time interval threshold to obtain the event trigger judgment result; according to the event trigger judgment result, the execution of trajectory complete re-planning, speed re-planning and other operations is selected. A mode of planning or not executing re-planning, while ensuring the stability of trajectory planning and high computational efficiency, has an adaptability similar to that of the planning method using a time-triggered mechanism in a dynamic and complex environment; and by building a complete set of abnormal events in the driving state of unmanned vehicles as the trigger conditions for planning, it can respond to and handle various types of autonomous driving scenarios, effectively reduce the number of unnecessary re-planning times, improve algorithm efficiency, reduce the computational burden of the computing platform, and improve the stability of trajectory planning and riding comfort, solving the technical problems in the existing technology that unmanned vehicles are too conservative in the face of dynamic obstacles, have low traffic efficiency, are difficult to cope with real complex road environments, and cannot take into account the safety and real-time nature of trajectory planning.

[0068] Reference Figure 1 , Figure 1 The figure is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention.

[0069] like Figure 1As shown, the device may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a Wi-Fi interface). The memory 1005 may be a high-speed RAM memory, or it may be a stable memory (Non-Volatile Memory), such as a disk memory. The memory 1005 may optionally be a storage device independent of the aforementioned processor 1001.

[0070] Those skilled in the art will understand that Figure 1 The device structure shown in the figure does not constitute a limitation of the device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0071] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating device, a network communication module, a user interface module, and an unmanned vehicle trajectory planning program.

[0072] The device of the present invention calls the unmanned vehicle trajectory planning program stored in the memory 1005 through the processor 1001, and performs the following operations:

[0073] Obtaining the current vehicle state information, environmental perception information, global map information, and last moment trajectory planning result information of the unmanned current vehicle, and performing a risk assessment on the surrounding traffic environment of the current vehicle based on the vehicle state information, the environmental perception information, the global map information, and the last moment trajectory planning result information to obtain a risk assessment result;

[0074] Calculate the time interval threshold of re-planning according to the risk assessment result, judge the event status in the preset planning trigger event set according to the time interval threshold, and obtain the event trigger judgment result;

[0075] A mode of executing complete trajectory replanning, speed replanning or not executing replanning is selected according to the event triggering judgment result.

[0076] The device of the present invention calls the unmanned vehicle trajectory planning program stored in the memory 1005 through the processor 1001, and also performs the following operations:

[0077] Obtain the current vehicle status information, environmental perception information, global map information and trajectory planning result information of the unmanned vehicle at the previous moment;

[0078] Determine the number of obstacles in a preset vehicle body area of ​​the current vehicle at a current vehicle speed, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles according to the vehicle state information, the environment perception information, the global map information, and the trajectory planning result information at the previous moment;

[0079] The risk assessment of the surrounding traffic environment of the current vehicle is performed according to the number of obstacles, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles to obtain a risk assessment result.

[0080] The device of the present invention calls the unmanned vehicle trajectory planning program stored in the memory 1005 through the processor 1001, and also performs the following operations:

[0081] According to the number of obstacles, the obstacle number risk sub-function is determined by the following formula:

[0082]

[0083] in, is the risk subfunction of the number of obstacles, is the number of obstacles within the current vehicle at the current moment, is the normalization coefficient;

[0084] According to the relative distance between the obstacle and the vehicle, the relative distance risk subfunction is determined by the following formula:

[0085]

[0086] in, is the relative distance risk subfunction, is the normalization coefficient, For the The lateral distance between the obstacle and the current vehicle, For the The longitudinal distance between the obstacle and the current vehicle, is the lateral distance weight coefficient, is the longitudinal distance weight coefficient;

[0087] According to the obstacle motion state, the motion state risk subfunction is determined by the following formula:

[0088]

[0089] in, is the motion state risk subfunction, is the normalization coefficient, is the obstacle motion state;

[0090] A traffic environment risk function is determined according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function, and a risk assessment is performed on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result.

[0091] The device of the present invention calls the unmanned vehicle trajectory planning program stored in the memory 1005 through the processor 1001, and also performs the following operations:

[0092] The traffic environment risk function is determined by the following formula according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function:

[0093]

[0094] in, is the traffic environment risk function, is the absolute time moment, is the obstacle quantity risk subfunction The weight coefficient of is the relative distance risk subfunction The weight coefficient of is the motion state risk subfunction The weight coefficient of

[0095] A risk assessment is performed on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result.

[0096] The device of the present invention calls the unmanned vehicle trajectory planning program stored in the memory 1005 through the processor 1001, and also performs the following operations:

[0097] The re-planning time interval threshold is calculated according to the risk assessment result by the following formula:

[0098]

[0099] in, is the re-planning time interval threshold, For preview duration, is the traffic environment risk function;

[0100] Obtain the start time corresponding to the start of the current plan and the time when the most recent trajectory replanning occurred from the preset plan trigger event set;

[0101] Compare the time difference between the start time and the occurrence time with the duration interval threshold, and when the time difference is greater than the duration interval threshold, determine that the event state is an event in which the time interval between the current moment and the last replanning is greater than the duration interval threshold;

[0102] Generate a first planning trigger event set according to the first planning start event, the time interval between the current moment and the last re-planning is greater than the time interval threshold event, the vehicle driving state switching event, and the distance deviation between the current position of the vehicle and the nearest point on the trajectory at the last moment is greater than the preset distance threshold event;

[0103] Taking the trajectory collision detection failure event as the second planning trigger event set;

[0104] An event status judgment is performed on the first planning triggering event set and the second planning triggering event set to obtain an event triggering judgment result.

[0105] The device of the present invention calls the unmanned vehicle trajectory planning program stored in the memory 1005 through the processor 1001, and also performs the following operations:

[0106] When at least one event in the first planning trigger event set occurs, the first trigger event is assigned a value of true, and the state judgment of the second planning trigger event set is terminated;

[0107] If none of the events in the first planned trigger event set occurs, the first trigger event is assigned a value of false, and the judgment of the second planned trigger event set is executed sequentially;

[0108] When at least one event in the second planned trigger event set occurs, assigning the second trigger event a value of true;

[0109] If none of the events in the second planned trigger event set occurs, the second trigger event is assigned a value of false;

[0110] Complete event trigger judgment and generate event trigger judgment results.

[0111] The device of the present invention calls the unmanned vehicle trajectory planning program stored in the memory 1005 through the processor 1001, and also performs the following operations:

[0112] When the event triggering judgment result is that both the first triggering event and the second triggering event are false, re-planning is not performed;

[0113] When the event triggering judgment result is that the first triggering event is true, performing trajectory complete replanning;

[0114] When the second trigger events are both true, speed replanning is performed.

[0115] Through the above scheme, this embodiment obtains the current vehicle state information, environmental perception information, global map information and trajectory planning result information of the unmanned current vehicle at the current moment, and performs risk assessment on the surrounding traffic environment of the current vehicle according to the vehicle state information, the environmental perception information, the global map information and the trajectory planning result information at the previous moment to obtain a risk assessment result; calculates the time interval threshold of re-planning according to the risk assessment result, and judges the event state in the preset planning trigger event set according to the time interval threshold to obtain an event trigger judgment result; selects one of the modes of executing trajectory complete re-planning, speed re-planning or not executing re-planning according to the event trigger judgment result, while ensuring the stability of trajectory planning and high calculation efficiency, it has an adaptability similar to that of the planning method using the time trigger mechanism in a dynamic and complex environment; and by constructing a complete set of abnormal events of the driving state of the unmanned vehicle as the trigger condition of planning, it can cope with and handle various automatic driving scenarios, effectively reduce the number of unnecessary re-planning times, improve algorithm efficiency, reduce the calculation burden of the computing platform, and improve the stability of trajectory planning and riding comfort.

[0116] Based on the above hardware structure, an embodiment of the unmanned vehicle trajectory planning method of the present invention is proposed.

[0117] Reference Figure 2 , Figure 2 Schematic diagram of the process of the first embodiment of the unmanned vehicle trajectory planning method of the present invention.

[0118] In a first embodiment, the unmanned vehicle trajectory planning method comprises the following steps:

[0119] Step S10: obtaining the current vehicle status information, environmental perception information, global map information and trajectory planning result information of the unmanned current vehicle, and performing a risk assessment on the surrounding traffic environment of the current vehicle based on the vehicle status information, the environmental perception information, the global map information and the trajectory planning result information of the previous moment to obtain a risk assessment result.

[0120] It should be noted that after obtaining the current vehicle status information, environmental perception information, global map information and trajectory planning result information of the unmanned vehicle at the previous moment, the risk of the traffic environment surrounding the current vehicle can be assessed based on the vehicle status information, the environmental perception information, the global map information and the trajectory planning result information at the previous moment to obtain a risk assessment result.

[0121] It should be understood that the vehicle status information includes the vehicle's current position, speed, acceleration, heading angle information, etc., and the vehicle's surrounding environment perception information includes obstacle information (for dynamic obstacles, it should include the predicted trajectory of the dynamic obstacles in the future), lane information and traffic sign information, etc.

[0122] Step S20: Calculate the time interval threshold of re-planning according to the risk assessment result, and judge the event status in the preset planning trigger event set according to the time interval threshold to obtain the event trigger judgment result.

[0123] It can be understood that the corresponding re-planning time interval threshold under the current environment is calculated based on the traffic risk assessment result, and then the event status of the pre-set preset planning trigger event set is judged according to the time interval threshold to obtain the event trigger judgment result.

[0124] Furthermore, the step S20 includes the following steps:

[0125] The re-planning time interval threshold is calculated according to the risk assessment result by the following formula:

[0126]

[0127] in, is the re-planning time interval threshold, For preview duration, is the traffic environment risk function;

[0128] Obtain the start time corresponding to the start of the current plan and the time when the most recent trajectory replanning occurred from the preset plan trigger event set;

[0129] Compare the time difference between the start time and the occurrence time with the duration interval threshold, and when the time difference is greater than the duration interval threshold, determine that the event state is an event in which the time interval between the current moment and the last replanning is greater than the duration interval threshold;

[0130] Generate a first planning trigger event set according to the first planning start event, the time interval between the current moment and the last re-planning is greater than the time interval threshold event, the vehicle driving state switching event, and the distance deviation between the current position of the vehicle and the nearest point on the trajectory at the last moment is greater than the preset distance threshold event;

[0131] Taking the trajectory collision detection failure event as the second planning trigger event set;

[0132] An event status judgment is performed on the first planning triggering event set and the second planning triggering event set to obtain an event triggering judgment result.

[0133] It should be noted that the above traffic environment risk R(t) ranges from [0 to 1]. The larger the R(t) value, the higher the traffic risk around the vehicle, and the greater the impact of obstacles on the vehicle's driving safety and traffic efficiency. Correspondingly, the time interval between two adjacent re-planning should be reduced, so the re-planning time interval threshold is

[0134] It is understandable that, Indicates the preview duration of a planning (the preview duration refers to the time range considered in a trajectory planning, which is generally a fixed value. For example, a planning process considers the vehicle's driving trajectory within the next 8 seconds). The value range is Through the traffic risk assessment process, the vehicle can dynamically adjust the time interval between two adjacent re-planning according to the real-time situation of surrounding obstacles.

[0135] In the specific implementation, in a dynamic and complex environment, the traffic environment risk is high, and the calculated time interval threshold between two re-plannings is small. At this time, the planning method shows a tendency to update the trajectory at a high frequency like a time trigger mechanism, which can enable the unmanned vehicle planning module to respond to potential risks in advance based on real-time environmental perception information and plan the optimal trajectory allowed by the algorithm used in the current environment; it improves the traditional event triggering that can only determine whether to execute re-planning based on the judgment result of the actual collision detection, resulting in the planning module being unable to respond to potential surrounding risks in a timely manner, overly conservative behavior, low driving efficiency, and even traffic accidents; in simple scenarios with lower risks, the traffic environment risk is low, and the time interval threshold between two re-plannings calculated by this model is also large. Through a larger upper limit on the re-planning time interval, the number of unnecessary re-planning times can be effectively reduced, the algorithm efficiency can be improved, the computational burden of the computing platform can be reduced, and the stability of trajectory planning and riding comfort can be improved.

[0136] It should be understood that the event status in the preset planning trigger event set is judged; the planning trigger event set is composed of a first planning trigger event set and a second planning trigger event set; the first planning trigger event set includes: the first planning is started, the time interval between the current moment and the last re-planning is greater than the re-planning time interval threshold, the vehicle's driving state is switched, and the distance deviation between the vehicle's current position and its nearest point on the trajectory at the last moment is greater than the distance threshold.

[0137] In the specific implementation, the first start of planning refers to the first time the trajectory planning algorithm runs, before which no historical trajectory planning results exist;

[0138] The calculation process of whether the time interval between the current time and the last replanning is greater than the replanning time interval threshold is as follows:

[0139]

[0140] In the formula, Indicates the time corresponding to the start of the current plan. Indicates the time when the most recent trajectory replanning occurred. The replanning interval threshold corresponding to the current environment.

[0141] The driving state switching of the autonomous vehicle refers to the switching of the autonomous vehicle between normal driving, braking, lane changing, parking, reversing and other working conditions in the autonomous lane. The driving state change can be controlled by a dedicated autonomous vehicle decision module or by an autonomous decision-making module; Figure 3 As shown, Figure 3 This is a schematic diagram of the vehicle position distance deviation in the unmanned vehicle trajectory planning method of the present invention, see Figure 3 When the distance deviation between the vehicle's current position and its nearest point on the trajectory at the previous moment is greater than the distance threshold, it indicates that the unmanned vehicle is not driving normally according to the planned trajectory at this time, and the state and safety of the unmanned vehicle are no longer controllable, and the trajectory planning needs to be re-performed.

[0142] It should be understood that the second planning trigger event set specifically refers to the trajectory collision detection failure event; this event is realized based on the collision detection judgment between the unmanned vehicle and the obstacle within the time and space range, in which the dynamic obstacle participates in the collision detection in the form of a predicted trajectory; Figure 4 FIG. 1 is a schematic diagram of the trajectory collision detection process in the unmanned vehicle trajectory planning method of the present invention, see Figure 4 Specifically, step 1, obtain the historical trajectory of the previous moment as the basic trajectory; step 2, calculate the closest point of the vehicle position on the basic trajectory at the current moment; step 3. Use the closest point as the segmentation point to divide the basic trajectory into the passed trajectory and the remaining valid trajectory; step 4, use the remaining valid trajectory as the current trajectory involved in collision detection; step 5, use the separating axis theorem or other collision detection methods to complete collision detection, and the specific method is not limited here.

[0143] Furthermore, the step of performing event status judgment on the first planning triggering event set and the second planning triggering event set to obtain an event triggering judgment result includes:

[0144] When at least one event in the first planning trigger event set occurs, the first trigger event is assigned a value of true, and the state judgment of the second planning trigger event set is terminated;

[0145] If none of the events in the first planned trigger event set occurs, the first trigger event is assigned a value of false, and the judgment of the second planned trigger event set is executed sequentially;

[0146] When at least one event in the second planned trigger event set occurs, assigning the second trigger event a value of true;

[0147] If none of the events in the second planned trigger event set occurs, the second trigger event is assigned a value of false;

[0148] Complete event trigger judgment and generate event trigger judgment results.

[0149] In the specific implementation, Figure 5 As shown, Figure 5 This is a schematic diagram of event state judgment in the unmanned vehicle trajectory planning method of the present invention, see Figure 5 , sequentially execute the event status judgment of the first planned trigger event set and the second planned trigger event set. If at least one event of the first planned trigger event set occurs, the first trigger event E1 is assigned to true, and the status judgment of the second planned trigger event set is terminated; conversely, if none of the events of the first planned trigger event set occurs, the first trigger event E1 is assigned to false, and the judgment of the second planned trigger event set is sequentially executed. If at least one event in the second planned trigger event set occurs, the second trigger event E2 is assigned to true, otherwise if none of the events in the second planned trigger event set occurs, the second trigger event E2 is assigned to false, and the event trigger judgment is completed.

[0150] Step S30: selecting a mode of performing complete trajectory replanning, speed replanning or not performing replanning according to the event triggering judgment result.

[0151] It should be understood that, according to the event triggering judgment result, one of three modes can be selected for execution, namely, trajectory complete replanning, speed replanning or no replanning.

[0152] Furthermore, the step S30 specifically includes the following steps:

[0153] When the event triggering judgment result is that both the first triggering event and the second triggering event are false, re-planning is not performed;

[0154] When the event triggering judgment result is that the first triggering event is true, performing trajectory complete replanning;

[0155] When the second trigger events are both true, speed replanning is performed.

[0156] It should be noted that the trajectory planning of unmanned vehicles generally includes two levels. The first level is path planning, which refers to planning a passable curve from the starting point to the end point at the spatial level; the second level is the speed part, that is, assigning speed to each path point of the unmanned vehicle; there are three modes of trajectory planning: A without re-planning, B speed planning, and C complete trajectory re-planning; among them, A does not re-plan, which means that the trajectory planning at the current moment does not need to be re-planned, and the trajectory planned last time can continue to be used as the result of the current trajectory planning; B speed planning, which means that the trajectory path part of the trajectory planning result at the previous moment does not need to be re-planned, and the trajectory speed part is re-planned based on this part of the path and the current information; C complete trajectory re-planning, which means that the trajectory planned at the previous moment is no longer available at the current moment, and the current trajectory needs to be re-planned.

[0157] In the specific implementation, Figure 5 As shown, if the first trigger event E1 and the second trigger event E2 are both false, the trajectory planning mode selection: A does not re-plan, and uses the trajectory planned at the previous moment as the trajectory planning result at the current moment to improve the stability of trajectory planning at adjacent moments, while avoiding unnecessary calculations and reducing the computing burden of the platform; if the first trigger event E1 is true, the trajectory planning mode selection: C complete trajectory re-planning, taking the current state of the vehicle as the starting point, and re-planning the drivable trajectory of the vehicle at the current moment; if the second trigger event E2 is true, indicating that the trajectory at the previous moment collides with the obstacle in the current environment, the trajectory planning mode first selects: B speed planning, and performs a collision check on the new trajectory generated after speed planning. If the new trajectory still cannot achieve collision avoidance, the trajectory planning mode selection: C complete trajectory re-planning, re-planning the vehicle trajectory to avoid obstacles or perform braking planning; thereby completing a trajectory planning.

[0158] Through the above scheme, this embodiment obtains the current vehicle state information, environmental perception information, global map information and trajectory planning result information of the unmanned current vehicle at the current moment, and performs risk assessment on the surrounding traffic environment of the current vehicle according to the vehicle state information, the environmental perception information, the global map information and the trajectory planning result information at the previous moment to obtain a risk assessment result; calculates the time interval threshold of re-planning according to the risk assessment result, and judges the event state in the preset planning trigger event set according to the time interval threshold to obtain an event trigger judgment result; selects one of the modes of executing trajectory complete re-planning, speed re-planning or not executing re-planning according to the event trigger judgment result, which can ensure the stability of trajectory planning and high calculation efficiency, and has an adaptability similar to that of the planning method using the time trigger mechanism in a dynamic and complex environment; and by constructing a complete set of abnormal events of the driving state of the unmanned vehicle as the trigger condition of planning, it can cope with and handle various automatic driving scenarios, effectively reduce the number of unnecessary re-planning times, improve algorithm efficiency, reduce the calculation burden of the computing platform, and improve the stability of trajectory planning and riding comfort.

[0159] Further, Figure 6 FIG. 2 is a flow chart of a second embodiment of the unmanned vehicle trajectory planning method of the present invention. Figure 6 As shown, a second embodiment of the unmanned vehicle trajectory planning method of the present invention is proposed based on the first embodiment. In this embodiment, the step S10 specifically includes the following steps:

[0160] Step S11, obtaining the current vehicle state information, environment perception information, global map information and trajectory planning result information of the unmanned vehicle at the previous moment.

[0161] It should be noted that the current vehicle status information, environmental perception information, global map information and trajectory planning result information of the unmanned vehicle at the previous moment are obtained. The vehicle status information includes the current position, speed, acceleration, heading angle information, etc. of the vehicle. The environmental perception information around the vehicle includes obstacle information (for dynamic obstacles, the predicted trajectory of the dynamic obstacles in the future should be included), lane information and traffic sign information, etc.

[0162] Step S12: determining the number of obstacles in a preset body area of ​​the current vehicle at a current speed, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles based on the vehicle state information, the environment perception information, the global map information, and the trajectory planning result information at the previous moment.

[0163] It can be understood that the number of obstacles in the preset body area of ​​the current vehicle at the current speed, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles are determined through the vehicle status information, the environmental perception information, the global map information and the trajectory planning result information at the previous moment.

[0164] In a specific implementation, the vehicle status information, the environmental perception information, the global map information and the trajectory planning result information at the previous moment can be used to determine the current speed of the vehicle and the number of obstacles in a certain area around the vehicle (the area can be defined as a potential risk area RR(t)), the relative distance between the obstacles and the vehicle, and the movement state of the obstacles. The current environment is risk modeled to assess the traffic risk around the vehicle, and the corresponding re-planning time interval threshold in the current environment is calculated based on the traffic risk assessment result.

[0165] Step S13: performing a risk assessment on the surrounding traffic environment of the current vehicle according to the number of obstacles, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles to obtain a risk assessment result.

[0166] It should be understood that the risk assessment result is obtained by performing a risk assessment on the surrounding traffic environment of the current vehicle through the number of obstacles, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles.

[0167] Furthermore, the step S13 specifically includes the following steps:

[0168] According to the number of obstacles, the obstacle number risk sub-function is determined by the following formula:

[0169]

[0170] in, is the risk subfunction of the number of obstacles, is the number of obstacles within the current vehicle at the current moment, is the normalization coefficient;

[0171] According to the relative distance between the obstacle and the vehicle, the relative distance risk subfunction is determined by the following formula:

[0172]

[0173] in, is the relative distance risk subfunction, is the normalization coefficient, For the The lateral distance between the obstacle and the current vehicle, For the The longitudinal distance between the obstacle and the current vehicle, is the lateral distance weight coefficient, is the longitudinal distance weight coefficient;

[0174] According to the obstacle motion state, the motion state risk subfunction is determined by the following formula:

[0175]

[0176] in, is the motion state risk subfunction, is the normalization coefficient, is the obstacle motion state;

[0177] A traffic environment risk function is determined according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function, and a risk assessment is performed on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result.

[0178] It should be noted that N(t) is the risk subfunction of the number of obstacles, where: and Respectively represent The lateral and longitudinal distances between the obstacle and the vehicle are calculated based on the Frenet coordinate system. and is the horizontal and vertical distance weight coefficient, is the normalization coefficient; where the obstacle motion includes three states: high speed, low speed and static. The corresponding motion values ​​are m1, m2 and m3. Since the greater the obstacle speed, the higher the risk to the vehicle, the value should satisfy m1>m2>m3. Represents the normalization coefficient.

[0179] Furthermore, the step of determining a traffic environment risk function according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function, and performing a risk assessment on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result comprises the following steps:

[0180] The traffic environment risk function is determined by the following formula according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function:

[0181]

[0182] in, is the traffic environment risk function, is the absolute time moment, is the obstacle quantity risk subfunction The weight coefficient of is the relative distance risk subfunction The weight coefficient of is the motion state risk subfunction The weight coefficient of

[0183] A risk assessment is performed on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result.

[0184] It can be understood that the re-planning time interval threshold refers to the maximum time interval allowed between two new trajectory planning. For the planning method using the time trigger mechanism, since each trajectory planning is brand new, the re-planning time interval is the time interval between two adjacent planning.

[0185] It should be understood that if Figure 7 As shown, Figure 7 Schematic diagram of real-time traffic environment risk modeling in the unmanned vehicle trajectory planning method of the present invention. The potential risk area RR(t) is L1m in front of the vehicle and L2m in the rear, and covers the range of the lane and adjacent lanes in the same direction in the lateral direction. L1 and L2 are related to the speed of the vehicle to dynamically adapt to changes in the vehicle state. The calculation process of the re-planning time interval threshold is shown in the above formula. The traffic environment risk function at the absolute time t; are weight coefficients corresponding to the above three sub-functions. Adjusting the three weight coefficients can make the risk function show different tendencies. The specific values ​​can be dynamically adjusted according to actual needs, and this embodiment does not limit this.

[0186] In the specific implementation, this embodiment determines the re-planning time interval threshold by modeling and evaluating the risk of the real-time traffic environment around the vehicle, so that the vehicle can dynamically adjust the time interval between two adjacent re-planning according to the real-time situation of obstacles in the surrounding environment; in a dynamic and complex environment, the traffic environment risk is high, and the calculated time interval between two re-plannings is short. At this time, the planning method shows a tendency to update the trajectory at a high frequency like a time trigger mechanism, which can enable the unmanned vehicle planning module to respond to potential risks in advance and plan the optimal trajectory allowed by the algorithm used in the current environment, thereby improving the traffic efficiency and driving safety of the unmanned vehicle in complex environments; it solves the existing planning method using an event trigger mechanism: since only the judgment result of the actual collision detection can be used to determine whether re-planning is needed, the planning module cannot respond to potential surrounding risks in a timely manner, the behavior is too conservative, the driving efficiency is low, and even traffic accidents may occur.

[0187] In simple scenarios with lower risks, the traffic environment risk is lower and the calculated time interval between two re-plannings is also longer. By setting a larger upper limit on the re-planning time interval, the number of unnecessary re-plannings can be effectively reduced, the algorithm efficiency can be improved, the computational burden of the computing platform can be reduced, and the stability of trajectory planning and riding comfort can be improved.

[0188] Based on the actual driving needs of unmanned vehicles, this embodiment constructs a complete set of abnormal driving state events of unmanned vehicles as trigger conditions for planning, which can cope with and handle various autonomous driving scenarios. In addition, in the trajectory planning step, based on the results of event trigger judgment, the setting and selection of three trajectory planning modes improves the efficiency and safety of the trajectory generated by the re-planning of the unmanned vehicle.

[0189] Through the above scheme, this embodiment obtains the current vehicle state information, environmental perception information, global map information and trajectory planning result information of the unmanned current vehicle at the previous moment; determines the number of obstacles in the preset body area of ​​the current vehicle at the current speed, the relative distance between the obstacles and the vehicle, and the obstacle motion state according to the vehicle state information, the environmental perception information, the global map information and the trajectory planning result information at the previous moment; performs risk assessment on the surrounding traffic environment of the current vehicle according to the number of obstacles, the relative distance between the obstacles and the vehicle, and the obstacle motion state, and obtains the risk assessment result. The risk assessment result can be accurately obtained, and while ensuring the stability and high calculation efficiency of trajectory planning, it has an adaptability similar to that of the planning method using the time trigger mechanism in a dynamic and complex environment.

[0190] Accordingly, the present invention further provides a trajectory planning device for an unmanned vehicle.

[0191] Reference Figure 8 , Figure 8 This is a functional module diagram of the first embodiment of the unmanned vehicle trajectory planning device of the present invention.

[0192] In a first embodiment of the unmanned vehicle trajectory planning device of the present invention, the unmanned vehicle trajectory planning device comprises:

[0193] The risk assessment module 10 is used to obtain the current vehicle status information, environmental perception information, global map information and trajectory planning result information of the unmanned current vehicle, and perform a risk assessment on the surrounding traffic environment of the current vehicle based on the vehicle status information, the environmental perception information, the global map information and the trajectory planning result information of the previous moment to obtain a risk assessment result.

[0194] The event trigger judgment module 20 is used to calculate the re-planning time interval threshold according to the risk assessment result, and judge the event status in the preset planning trigger event set according to the time interval threshold to obtain the event trigger judgment result.

[0195] The planning execution module 30 is used to select a mode of executing trajectory complete replanning, speed replanning or not executing replanning according to the event triggering judgment result.

[0196] The risk assessment module 10 is also used to obtain the current vehicle status information, environmental perception information, global map information and trajectory planning result information of the unmanned current vehicle; determine the number of obstacles in a preset body area of ​​the current vehicle at the current speed, the relative distance between the obstacles and the vehicle, and the obstacle motion state according to the vehicle status information, the environmental perception information, the global map information and the trajectory planning result information of the previous moment; and perform a risk assessment on the surrounding traffic environment of the current vehicle according to the number of obstacles, the relative distance between the obstacles and the vehicle, and the obstacle motion state to obtain a risk assessment result.

[0197] The risk assessment module 10 is further used to determine the obstacle quantity risk sub-function according to the obstacle quantity by the following formula:

[0198]

[0199] in, is the risk subfunction of the number of obstacles, is the number of obstacles within the current vehicle at the current moment, is the normalization coefficient;

[0200] According to the relative distance between the obstacle and the vehicle, the relative distance risk subfunction is determined by the following formula:

[0201]

[0202] in, is the relative distance risk subfunction, is the normalization coefficient, For the The lateral distance between the obstacle and the current vehicle, For the The longitudinal distance between the obstacle and the current vehicle, is the lateral distance weight coefficient, is the longitudinal distance weight coefficient;

[0203] According to the obstacle motion state, the motion state risk subfunction is determined by the following formula:

[0204]

[0205] in, is the motion state risk subfunction, is the normalization coefficient, is the obstacle motion state;

[0206] A traffic environment risk function is determined according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function, and a risk assessment is performed on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result.

[0207] The risk assessment module 10 is further used to determine the traffic environment risk function according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function by the following formula:

[0208]

[0209] in, is the traffic environment risk function, is the absolute time moment, is the obstacle quantity risk subfunction The weight coefficient of is the relative distance risk subfunction The weight coefficient of is the motion state risk subfunction The weight coefficient of

[0210] A risk assessment is performed on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result.

[0211] The event triggering judgment module 20 is further used to calculate the re-planning time interval threshold according to the risk assessment result by the following formula:

[0212]

[0213] in, is the re-planning time interval threshold, For preview duration, is the traffic environment risk function;

[0214] Obtain the start time corresponding to the start of the current plan and the time when the most recent trajectory replanning occurred from the preset plan trigger event set;

[0215] Compare the time difference between the start time and the occurrence time with the duration interval threshold, and when the time difference is greater than the duration interval threshold, determine that the event state is an event in which the time interval between the current moment and the last replanning is greater than the duration interval threshold;

[0216] Generate a first planning trigger event set according to the first planning start event, the time interval between the current moment and the last re-planning is greater than the time interval threshold event, the vehicle driving state switching event, and the distance deviation between the current position of the vehicle and the nearest point on the trajectory at the last moment is greater than the preset distance threshold event;

[0217] Taking the trajectory collision detection failure event as the second planning trigger event set;

[0218] An event status judgment is performed on the first planning triggering event set and the second planning triggering event set to obtain an event triggering judgment result.

[0219] The event trigger judgment module 20 is also used to, when at least one event in the first planned trigger event set occurs, assign the first trigger event to true and terminate the state judgment of the second planned trigger event set; when none of the events in the first planned trigger event set occurs, assign the first trigger event to false and sequentially execute the judgment of the second planned trigger event set; when at least one event in the second planned trigger event set occurs, assign the second trigger event to true; when none of the events in the second planned trigger event set occurs, assign the second trigger event to false; complete the event trigger judgment and generate an event trigger judgment result.

[0220] The planning execution module 30 is also used for not executing replanning when the event trigger judgment result is that both the first trigger event and the second trigger event are false; executing trajectory complete replanning when the event trigger judgment result is that the first trigger event is true; and executing speed replanning when both the second trigger events are true.

[0221] Among them, the steps implemented by each functional module of the unmanned vehicle trajectory planning device can refer to the various embodiments of the unmanned vehicle trajectory planning method of the present invention, and will not be repeated here.

[0222] In addition, an embodiment of the present invention further provides a storage medium, on which an unmanned vehicle trajectory planning program is stored, and when the unmanned vehicle trajectory planning program is executed by a processor, the following operations are implemented:

[0223] Obtaining the current vehicle state information, environmental perception information, global map information, and last moment trajectory planning result information of the unmanned current vehicle, and performing a risk assessment on the surrounding traffic environment of the current vehicle based on the vehicle state information, the environmental perception information, the global map information, and the last moment trajectory planning result information to obtain a risk assessment result;

[0224] Calculate the time interval threshold of re-planning according to the risk assessment result, judge the event status in the preset planning trigger event set according to the time interval threshold, and obtain the event trigger judgment result;

[0225] A mode of executing complete trajectory replanning, speed replanning or not executing replanning is selected according to the event triggering judgment result.

[0226] Furthermore, when the unmanned vehicle trajectory planning program is executed by the processor, the following operations are also implemented:

[0227] Obtain the current vehicle status information, environmental perception information, global map information and trajectory planning result information of the unmanned vehicle at the previous moment;

[0228] Determine the number of obstacles in a preset vehicle body area of ​​the current vehicle at a current vehicle speed, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles according to the vehicle state information, the environment perception information, the global map information, and the trajectory planning result information at the previous moment;

[0229] The risk assessment of the surrounding traffic environment of the current vehicle is performed according to the number of obstacles, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles to obtain a risk assessment result.

[0230] Furthermore, when the unmanned vehicle trajectory planning program is executed by the processor, the following operations are also implemented:

[0231] According to the number of obstacles, the obstacle number risk sub-function is determined by the following formula:

[0232]

[0233] in, is the risk subfunction of the number of obstacles, is the number of obstacles within the current vehicle at the current moment, is the normalization coefficient;

[0234] According to the relative distance between the obstacle and the vehicle, the relative distance risk subfunction is determined by the following formula:

[0235]

[0236] in, is the relative distance risk subfunction, is the normalization coefficient, For the The lateral distance between the obstacle and the current vehicle, For the The longitudinal distance between the obstacle and the current vehicle, is the lateral distance weight coefficient, is the longitudinal distance weight coefficient;

[0237] According to the obstacle motion state, the motion state risk subfunction is determined by the following formula:

[0238]

[0239] in, is the motion state risk subfunction, is the normalization coefficient, is the obstacle motion state;

[0240] A traffic environment risk function is determined according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function, and a risk assessment is performed on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result.

[0241] Furthermore, when the unmanned vehicle trajectory planning program is executed by the processor, the following operations are also implemented:

[0242] The traffic environment risk function is determined by the following formula according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function:

[0243]

[0244] in, is the traffic environment risk function, is the absolute time moment, is the obstacle quantity risk subfunction The weight coefficient of is the relative distance risk subfunction The weight coefficient of is the motion state risk subfunction The weight coefficient of

[0245] A risk assessment is performed on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result.

[0246] Furthermore, when the unmanned vehicle trajectory planning program is executed by the processor, the following operations are also implemented:

[0247] The re-planning time interval threshold is calculated according to the risk assessment result by the following formula:

[0248]

[0249] in, is the re-planning time interval threshold, For preview duration, is the traffic environment risk function;

[0250] Obtain the start time corresponding to the start of the current plan and the time when the most recent trajectory replanning occurred from the preset plan trigger event set;

[0251] Compare the time difference between the start time and the occurrence time with the duration interval threshold, and when the time difference is greater than the duration interval threshold, determine that the event state is an event in which the time interval between the current moment and the last replanning is greater than the duration interval threshold;

[0252] Generate a first planning trigger event set according to the first planning start event, the time interval between the current moment and the last re-planning is greater than the time interval threshold event, the vehicle driving state switching event, and the distance deviation between the current position of the vehicle and the nearest point on the trajectory at the last moment is greater than the preset distance threshold event;

[0253] Taking the trajectory collision detection failure event as the second planning trigger event set;

[0254] An event status judgment is performed on the first planning triggering event set and the second planning triggering event set to obtain an event triggering judgment result.

[0255] Furthermore, when the unmanned vehicle trajectory planning program is executed by the processor, the following operations are also implemented:

[0256] When at least one event in the first planning trigger event set occurs, the first trigger event is assigned a value of true, and the state judgment of the second planning trigger event set is terminated;

[0257] If none of the events in the first planned trigger event set occurs, the first trigger event is assigned a value of false, and the judgment of the second planned trigger event set is executed sequentially;

[0258] When at least one event in the second planned trigger event set occurs, assigning the second trigger event a value of true;

[0259] If none of the events in the second planned trigger event set occurs, the second trigger event is assigned a value of false;

[0260] Complete event trigger judgment and generate event trigger judgment results.

[0261] Furthermore, when the unmanned vehicle trajectory planning program is executed by the processor, the following operations are also implemented:

[0262] When the event triggering judgment result is that both the first triggering event and the second triggering event are false, re-planning is not performed;

[0263] When the event triggering judgment result is that the first triggering event is true, performing trajectory complete replanning;

[0264] When the second trigger events are both true, speed replanning is performed.

[0265] Those skilled in the art will appreciate that all or part of the steps in the above-mentioned implementation method can be accomplished by instructing related hardware through a program, and the program is stored in a storage medium, including a number of instructions for enabling a device (which may be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the method described in each implementation method of the present application; and the aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0266] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0267] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0268] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for trajectory planning of an unmanned vehicle, characterized in that: The unmanned vehicle trajectory planning method comprises: Obtaining the current vehicle state information, environmental perception information, global map information, and last moment trajectory planning result information of the unmanned current vehicle, and performing a risk assessment on the surrounding traffic environment of the current vehicle based on the vehicle state information, the environmental perception information, the global map information, and the last moment trajectory planning result information to obtain a risk assessment result; Calculate the time interval threshold of re-planning according to the risk assessment result, judge the event status in the preset planning trigger event set according to the time interval threshold, and obtain the event trigger judgment result; According to the event trigger judgment result, one of the following modes is selected: trajectory complete replanning, speed replanning, or no replanning; The step of calculating the time interval threshold of re-planning according to the risk assessment result, judging the event status in the preset planning trigger event set according to the time interval threshold, and obtaining the event trigger judgment result includes: The re-planning time interval threshold is calculated according to the risk assessment result by the following formula: in, is the re-planning time interval threshold, For preview duration, is the traffic environment risk function; Obtain the start time corresponding to the start of the current plan and the time when the most recent trajectory replanning occurred from the preset plan trigger event set; Compare the time difference between the start time and the occurrence time with the duration interval threshold, and when the time difference is greater than the duration interval threshold, determine that the event state is an event in which the time interval between the current moment and the last replanning is greater than the duration interval threshold; Generate a first planning trigger event set according to the first planning start event, the time interval between the current moment and the last re-planning is greater than the time interval threshold event, the vehicle driving state switching event, and the distance deviation between the current position of the vehicle and the nearest point on the trajectory at the last moment is greater than the preset distance threshold event; Taking the trajectory collision detection failure event as the second planning trigger event set; An event status judgment is performed on the first planning triggering event set and the second planning triggering event set to obtain an event triggering judgment result.

2. The unmanned vehicle trajectory planning method according to claim 1, characterized in that: The step of obtaining the current state information, environmental perception information, global map information, and trajectory planning result information of the unmanned current vehicle, and performing risk assessment on the surrounding traffic environment of the current vehicle according to the vehicle state information, the environmental perception information, the global map information, and the trajectory planning result information of the previous moment to obtain a risk assessment result includes: Obtain the current vehicle status information, environmental perception information, global map information and trajectory planning result information of the unmanned vehicle at the previous moment; Determine the number of obstacles in a preset vehicle body area of ​​the current vehicle at a current vehicle speed, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles according to the vehicle state information, the environment perception information, the global map information, and the trajectory planning result information at the previous moment; The risk assessment of the surrounding traffic environment of the current vehicle is performed according to the number of obstacles, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles to obtain a risk assessment result.

3. The unmanned vehicle trajectory planning method according to claim 2, characterized in that: The risk assessment of the surrounding traffic environment of the current vehicle is performed according to the number of obstacles, the relative distance between the obstacles and the vehicle, and the movement state of the obstacles to obtain a risk assessment result, including: According to the number of obstacles, the obstacle number risk sub-function is determined by the following formula: in, is the risk subfunction of the number of obstacles, is the number of obstacles within the current vehicle at the current moment, is the normalization coefficient; According to the relative distance between the obstacle and the vehicle, the relative distance risk subfunction is determined by the following formula: in, is the relative distance risk subfunction, is the normalization coefficient, For the The lateral distance between the obstacle and the current vehicle, For the The longitudinal distance between the obstacle and the current vehicle, is the lateral distance weight coefficient, is the longitudinal distance weight coefficient; According to the obstacle motion state, the motion state risk subfunction is determined by the following formula: in, is the motion state risk subfunction, is the normalization coefficient, is the obstacle motion state; A traffic environment risk function is determined according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function, and a risk assessment is performed on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result.

4. The unmanned vehicle trajectory planning method according to claim 3, characterized in that: The determining of the traffic environment risk function according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function, and performing risk assessment on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result includes: The traffic environment risk function is determined by the following formula according to the obstacle quantity risk sub-function, the relative distance risk sub-function and the motion state risk sub-function: in, is the traffic environment risk function, is the absolute time moment, is the obstacle quantity risk subfunction The weight coefficient of is the relative distance risk subfunction The weight coefficient of is the motion state risk subfunction The weight coefficient of A risk assessment is performed on the surrounding traffic environment of the current vehicle according to the traffic environment risk function to obtain a risk assessment result.

5. The unmanned vehicle trajectory planning method according to claim 1, characterized in that: The performing event status judgment on the first planning triggering event set and the second planning triggering event set to obtain an event triggering judgment result includes: When at least one event in the first planned trigger event set occurs, the first trigger event is assigned a value of true, and the state judgment of the second planned trigger event set is terminated; If none of the events in the first planned trigger event set occurs, the first trigger event is assigned a value of false, and the judgment of the second planned trigger event set is executed sequentially; When at least one event in the second planned trigger event set occurs, assigning a second trigger event value to true; If none of the events in the second planned trigger event set occurs, the second trigger event is assigned a value of false; Complete event trigger judgment and generate event trigger judgment results.

6. The unmanned vehicle trajectory planning method according to claim 5, characterized in that: The selecting, according to the event triggering judgment result, to execute one of the modes of trajectory complete replanning, speed replanning or not executing replanning comprises: When the event triggering judgment result is that both the first triggering event and the second triggering event are false, re-planning is not performed; When the event triggering judgment result is that the first triggering event is true, performing trajectory complete replanning; When the second trigger events are both true, speed replanning is performed.

7. An unmanned vehicle trajectory planning device, characterized in that: The unmanned vehicle trajectory planning device comprises: A risk assessment module is used to obtain the current vehicle state information, environmental perception information, global map information and trajectory planning result information of the unmanned current vehicle, and to perform a risk assessment on the surrounding traffic environment of the current vehicle based on the vehicle state information, the environmental perception information, the global map information and the trajectory planning result information of the previous moment to obtain a risk assessment result; An event trigger judgment module is used to calculate a time interval threshold for re-planning according to the risk assessment result, and to judge the event status in the preset planning trigger event set according to the time interval threshold to obtain an event trigger judgment result; A planning execution module, used for selecting a mode of executing trajectory complete replanning, speed replanning or not executing replanning according to the event trigger judgment result; The event triggering judgment module is further used to calculate the re-planning time interval threshold according to the risk assessment result by the following formula: in, is the re-planning time interval threshold, For preview duration, is the traffic environment risk function; Obtain the start time corresponding to the start of the current plan and the time when the most recent trajectory replanning occurred from the preset plan trigger event set; Compare the time difference between the start time and the occurrence time with the duration interval threshold, and when the time difference is greater than the duration interval threshold, determine that the event state is an event in which the time interval between the current moment and the last replanning is greater than the duration interval threshold; Generate a first planning trigger event set according to the first planning start event, the time interval between the current moment and the last re-planning is greater than the time interval threshold event, the vehicle driving state switching event, and the distance deviation between the current position of the vehicle and the nearest point on the trajectory at the last moment is greater than the preset distance threshold event; Taking the trajectory collision detection failure event as the second planning trigger event set; An event status judgment is performed on the first planning triggering event set and the second planning triggering event set to obtain an event triggering judgment result.

8. An unmanned vehicle trajectory planning device, characterized in that: The unmanned vehicle trajectory planning device includes: a memory, a processor, and an unmanned vehicle trajectory planning program stored in the memory and executable on the processor, wherein the unmanned vehicle trajectory planning program is configured to implement the steps of the unmanned vehicle trajectory planning method as described in any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium stores an unmanned vehicle trajectory planning program, and when the unmanned vehicle trajectory planning program is executed by the processor, the steps of the unmanned vehicle trajectory planning method according to any one of claims 1 to 6 are implemented.

Images