Unmanned vehicle automatic platooning system and method

Abstract

The present invention relates to the field of automatic driving technology, and provides an unmanned vehicle automatic formation system and method, a data acquisition module, used to obtain the driving environment and the number of slave vehicles of the unmanned vehicle; a driving formation determination module, used to determine the driving formation of the unmanned vehicle according to the relative position of the slave vehicle in the formation formation and the navigation trajectory data of the master vehicle; a formation recovery module, used to identify the formation position of the task vehicle, and set the formation recovery rules of the unmanned vehicle in combination with the trajectory prediction network of the side vehicle; a predicament perception module, used to perceive the vehicle driving state and environmental change trend of the unmanned vehicle, and set the predicament perception mode of the unmanned vehicle; a single vehicle escape module, used to identify the abnormal situation of the unmanned vehicle, and set the single vehicle escape mechanism of the unmanned vehicle; an automatic formation module, used for the automatic formation processing of the unmanned vehicle, and obtain the automatic formation result. The present invention is to improve the self-escaping ability of the unmanned vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of computer unmanned vehicle automatic platooning, and in particular to an unmanned vehicle automatic platooning system and method. Background Art

[0002] Automatic platooning of unmanned vehicles refers to the process in which multiple autonomous vehicles automatically switch driving formations according to the driving environment and travel in coordination in a certain queue order in the field of autonomous driving. With the rapid development of artificial intelligence and 5G mobile communications, autonomous driving technology has become an important development direction of the automotive industry. As an important branch of autonomous driving technology, unmanned vehicle platooning can bring many advantages. However, in actual applications, platoon vehicles may encounter various complex environments and emergencies, causing the unmanned vehicle platoon to fall into trouble. Therefore, how to effectively solve the abnormal situations of unmanned vehicle platoons is particularly important.

[0003] In the existing technology, the leading and following method is mainly used to control the formation of unmanned vehicles. Although this method can improve the vehicle driving efficiency to a certain extent, if the unmanned vehicle encounters an abnormal situation during autonomous driving, it may cause the vehicle to be unable to escape on its own, affecting the mission execution efficiency of the unmanned vehicle. Summary of the invention

[0004] The present invention provides an unmanned vehicle automatic formation system and method, the main purpose of which is to improve the unmanned vehicle's ability to escape from trouble on its own.

[0005] To achieve the above-mentioned purpose, the present invention provides an unmanned vehicle automatic formation system, comprising: a data acquisition module, a driving formation determination module, a formation recovery module, a predicament perception module, a single vehicle escape module, and an automatic formation module;

[0006] The data acquisition module is used to acquire the unmanned vehicles to be automatically formed, determine the master vehicle and the slave vehicles of the unmanned vehicles, collect the navigation trajectory data of the master vehicle, and identify the driving environment of the unmanned vehicle and the number of the slave vehicles according to the navigation trajectory data;

[0007] The driving formation determination module is used to set the formation of the unmanned vehicle based on the driving environment and the number of the slave vehicles, locate the relative position of the slave vehicle in the formation, and determine the driving formation of the unmanned vehicle based on the navigation trajectory data and the relative position;

[0008] The formation recovery module is used to identify the task vehicle in the driving formation and the formation position of the task vehicle, build a side vehicle trajectory prediction network of the task vehicle based on the formation position, and set the formation recovery rules of the unmanned vehicle according to the side vehicle trajectory prediction network and the formation position;

[0009] The distress perception module is used to perceive the driving state of the unmanned vehicle in real time in combination with the driving formation and the formation recovery rule, analyze the environmental change trend of the driving environment based on the navigation trajectory data, and set the distress perception mode of the unmanned vehicle in combination with the vehicle driving state and the environmental change trend;

[0010] The single vehicle escape module is used to identify the abnormal condition of the unmanned vehicle based on the distress perception mode; and set the single vehicle escape mechanism of the unmanned vehicle according to the abnormal condition of the vehicle and the adjacent vehicle trajectory prediction network;

[0011] The automatic formation module is used to automatically form a formation for the unmanned vehicles according to the distress perception mode, the single-vehicle escape mechanism and the formation recovery rule to obtain an automatic formation result.

[0012] Optionally, determining the driving formation of the unmanned vehicle based on the navigation trajectory data and the relative position includes:

[0013] Based on the navigation trajectory data, identifying the driving trajectory of the unmanned vehicle;

[0014] According to the driving trajectory, dispatching the driving road condition information and formation of the unmanned vehicle;

[0015] Based on the driving road condition information and the relative position, adaptively adjusting the formation to obtain an adaptive formation;

[0016] The driving formation of the unmanned vehicle is determined by combining the driving road condition information and the adaptive formation.

[0017] Optionally, constructing a trajectory prediction network of a vehicle beside the task vehicle based on the formation position includes:

[0018] Based on the formation position, calculating the relative position of the task vehicle and the adjacent vehicles;

[0019] Identifying a non-mission vehicle at a relative position of the adjacent vehicle;

[0020] analyzing characteristic behaviors of the non-mission vehicle relative to the mission vehicle;

[0021] Extracting the driving behavior of the task vehicle, and analyzing the driving impact of the driving behavior on the non-task vehicle based on the characteristic behavior;

[0022] According to the characteristic behavior and the driving impact, a vehicle trajectory prediction network of the task vehicle is constructed.

[0023] Optionally, setting the formation recovery rule of the unmanned vehicle according to the side vehicle trajectory prediction network and the formation position includes:

[0024] Monitoring the vehicles in the convoy of the unmanned vehicles in real time and identifying the vehicles that have left the convoy;

[0025] Analyzing the impact of the departing vehicle on the formation of the unmanned vehicle according to the adjacent vehicle trajectory prediction network;

[0026] Dispatching communication connection data between the unmanned vehicle and the leaving vehicle, and identifying the leaving reason of the leaving vehicle based on the communication connection data;

[0027] According to the formation impact and the reason for leaving the team, a vehicle compensation mechanism is set for the leaving vehicle;

[0028] Identify the driving trajectory of the unmanned vehicle, and set a path for the vehicle leaving the team to return to the team based on the driving trajectory and the vehicle compensation mechanism;

[0029] In combination with the formation position and the re-entry path, a formation recovery rule for the unmanned vehicle is set.

[0030] Optionally, analyzing the environmental change trend of the driving environment based on the navigation trajectory data includes:

[0031] Attribute classification is performed on the navigation trajectory data to obtain classification data;

[0032] Extracting key features of the classified data, and collecting historical data of the driving environment based on the key features;

[0033] identifying traffic data and weather data in the historical data;

[0034] Analyzing the traffic flow and vehicle behavior patterns of the driving environment based on the traffic data;

[0035] analyzing a visibility level of the driving environment based on the weather data;

[0036] The environmental change trend of the driving environment is analyzed in combination with the traffic flow, the vehicle behavior pattern and the visibility level.

[0037] Optionally, the setting of the distress perception mode of the unmanned vehicle in combination with the vehicle driving state and the environmental change trend includes:

[0038] Based on the driving state of the vehicle, identifying vehicles in the surrounding environment of the unmanned vehicle;

[0039] Analyzing the interaction relationship between the surrounding environment vehicles and the unmanned vehicle;

[0040] Identifying potential risk factors of the unmanned vehicle according to the environmental change trend and the interactive relationship, and identifying risk levels of the potential risk factors;

[0041] Setting a risk response mode of the unmanned vehicle according to the vehicle driving state and the risk level;

[0042] In combination with the risk level and the risk response method, a distress perception mode of the unmanned vehicle is set.

[0043] Optionally, the identifying an abnormal situation of the unmanned vehicle based on the distress perception mode includes:

[0044] Based on the dilemma perception mode, collecting historical driving data of the unmanned vehicle;

[0045] Identifying normal driving parameters of the unmanned vehicle based on the historical driving data;

[0046] Identifying a current driving state of the unmanned vehicle, and identifying a current behavior deviation of the unmanned vehicle by combining the current driving state with the historical driving data;

[0047] Based on the normal driving parameters, setting an abnormal parameter threshold of the unmanned vehicle;

[0048] According to the abnormal parameter threshold and the current behavior deviation, an abnormal condition of the unmanned vehicle is identified.

[0049] Optionally, the step of setting a single vehicle escape mechanism for the unmanned vehicle according to the abnormal condition of the vehicle and the adjacent vehicle trajectory prediction network includes:

[0050] According to the abnormal condition of the vehicle, locate the abnormal single vehicle of the unmanned vehicle;

[0051] Analyze the dilemma type of the abnormal bicycle, and set a dilemma response layer for the abnormal bicycle based on the dilemma type;

[0052] According to the side vehicle trajectory prediction network, identifying the side vehicle driving trajectory of the abnormal single vehicle;

[0053] Based on the driving trajectory of the adjacent vehicle, setting an escape path for the abnormal single vehicle;

[0054] In combination with the distress response layer and the escape path, a single-vehicle escape mechanism for the unmanned vehicle is set.

[0055] Optionally, the automatic formation processing of the unmanned vehicles is performed according to the distress perception mode, the single-vehicle escape mechanism and the formation recovery rule to obtain the automatic formation result, including:

[0056] identifying a driving difficulty of the unmanned vehicle according to the difficulty perception mode;

[0057] Based on the vehicle driving predicament, setting the triggering conditions of the single-vehicle escape mechanism and setting the risk avoidance method of the unmanned vehicle;

[0058] According to the trigger condition and the risk avoidance method, the unmanned vehicle is rescued to obtain a rescue result;

[0059] Based on the escape result and the formation recovery rule, the unmanned vehicles are automatically formed into a team to obtain an automatic formation result.

[0060] The present invention collects the navigation trajectory data of the master vehicle, identifies the driving environment and the number of slave vehicles of the unmanned vehicle according to the navigation trajectory data, and can clarify the roles of the master vehicle and the slave vehicles, thereby realizing efficient driving of the unmanned vehicle formation; secondly, the embodiment of the present invention sets the formation of the unmanned vehicle based on the driving environment and the number of slave vehicles, and can judge the relative position of the slave vehicle in the formation, thereby determining the final driving formation, reducing the air resistance of the unmanned vehicle during driving, and improving driving efficiency and vehicle safety; thirdly, the embodiment of the present invention sets the formation recovery rules of the unmanned vehicle according to the side vehicle trajectory prediction network and the formation position, so as to ensure that the unmanned vehicle can adapt to the changes in the driving environment at any time during driving and maintain the formation formation; further, the embodiment of the present invention perceives the vehicle driving state of the unmanned vehicle in real time by combining the driving formation and the formation recovery rules, and can obtain the real-time operation status of the task vehicle and the non-task vehicle, timely discover potential safety problems, and improve safety. At the same time, combined with the vehicle driving state and the environmental change trend, the dilemma perception mode of the unmanned vehicle is set to predict potential dangerous situations and adjust the formation vehicles in time. The driving strategy can enhance the robustness of the unmanned vehicle system in the face of complex and unpredictable environments; the embodiment of the present invention can identify the abnormal situation of the unmanned vehicle based on the distress perception mode, and set the single-vehicle escape mechanism of the unmanned vehicle according to the abnormal situation of the vehicle and the trajectory prediction network of the adjacent vehicle, so as to ensure that each vehicle in the unmanned vehicle can operate normally, which is helpful to maintain the continuity and stability of the unmanned vehicle formation, and ensure that the abnormal vehicle can safely escape from the distress, improve the adaptability of the unmanned vehicle system in the face of emergencies, and ensure the stability and reliability of the entire traffic flow; finally, the embodiment of the present invention can effectively respond to external interference, maintain the stability of the unmanned vehicle formation, and dynamically adjust the formation structure and control strategy to adapt to environmental changes. The robustness of the unmanned vehicle system in the face of complex and changeable environments is enhanced. At the same time, through effective distress perception and single-vehicle escape, the unmanned vehicle formation can reduce unnecessary parking and deceleration caused by avoiding obstacles or handling emergencies, which can ensure the continuous execution of tasks and reduce task delays. Therefore, the unmanned vehicle automatic platooning system and method provided by the embodiments of the present invention can improve the unmanned vehicle's ability to escape from trouble on its own. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A functional module diagram of an unmanned vehicle automatic platooning system provided by one embodiment of the present invention;

[0062] Figure 2 A schematic flow chart of an automatic platooning method for unmanned vehicles provided in accordance with an embodiment of the present invention.

[0063] 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

[0064] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] Reference Figure 1 , which is a functional module diagram of an unmanned vehicle automatic platooning system provided by one embodiment of the present invention.

[0066] The unmanned vehicle automatic formation system 100 of the present invention can be set in a cloud server. In terms of implementation, it can be used as one or more service devices, or as an application installed on the cloud (such as a server or server cluster of a live broadcast service operator), or it can also be developed as a website. According to the functions implemented, the unmanned vehicle automatic formation system 100 includes a data acquisition module 101, a driving formation determination module 102, a formation recovery module 103, a predicament perception module 104, a single vehicle escape module 105 and an automatic formation module 106.

[0067] In the embodiment of the present invention, in the tracking based on the automatic formation of unmanned vehicles, each of the above modules can be implemented independently and called with other modules. The call here can be understood as a module that can connect multiple modules of another type and provide corresponding services to the multiple modules connected to it. In the unmanned vehicle automatic formation system provided by the embodiment of the present invention, there is no need to modify the program code. The scope of application of the unmanned vehicle automatic formation architecture can be adjusted by adding modules and directly calling them to achieve cluster-type horizontal expansion, so as to achieve the purpose of quickly and flexibly expanding the unmanned vehicle automatic formation system. In practical applications, the above modules can be set in the same device or different devices, or they can be set in virtual devices, such as service instances in cloud servers.

[0068] The following describes the various components and specific workflow of the unmanned vehicle automatic platooning system in conjunction with specific embodiments.

[0069] The data acquisition module 101 is used to obtain the unmanned vehicles to be automatically formed, determine the master vehicle and the slave vehicles of the unmanned vehicles, collect the navigation trajectory data of the master vehicle, and identify the driving environment of the unmanned vehicle and the number of slave vehicles based on the navigation trajectory data.

[0070] The embodiment of the present invention obtains the unmanned vehicles to be automatically formed into a platoon and determines the master vehicle and the slave vehicle of the unmanned vehicles, so as to clarify the roles of the master vehicle and the slave vehicle and realize efficient driving of the unmanned vehicle platoon. The unmanned vehicles refer to a traffic flow composed of multiple self-driving vehicles, the master vehicle refers to the leading vehicle among the unmanned vehicles, and the slave vehicle refers to the vehicle following the master vehicle among the unmanned vehicles.

[0071] Optionally, the master vehicle and the slave vehicle of the unmanned vehicle may be determined by utilizing vehicle recognition technology, such as license plate recognition technology.

[0072] Furthermore, the embodiments of the present invention can provide data support for subsequent unmanned vehicle driving by collecting the navigation trajectory data of the main vehicle. The navigation trajectory data refers to a data set of the position, speed and other relevant driving information on the predetermined path or route followed by the leading vehicle in the unmanned vehicle, such as dynamic obstacle information, navigation instructions and environmental perception data.

[0073] Optionally, the navigation trajectory data of the host vehicle may be acquired through a vehicle navigation system.

[0074] The embodiment of the present invention can analyze and obtain the most efficient driving formation of the unmanned vehicle under the current environment by identifying the driving environment and the number of follower vehicles of the unmanned vehicle based on the navigation trajectory data. The driving environment refers to the external conditions and surrounding environment of the unmanned vehicle, and the number of follower vehicles refers to the number of vehicles following the main vehicle in the formation.

[0075] Optionally, based on the navigation trajectory data, the driving environment identification of the unmanned vehicle can be obtained through vehicle-mounted sensors, such as radar, and based on the navigation trajectory data, the number of subordinate vehicles of the unmanned vehicle can be identified using a wireless communication network, such as a 5G network.

[0076] The driving formation determination module 102 is used to set the formation of the unmanned vehicle based on the driving environment and the number of the slave vehicles, locate the relative position of the slave vehicles in the formation, and determine the driving formation of the unmanned vehicle based on the navigation trajectory data and the relative position.

[0077] The embodiment of the present invention sets the formation of the unmanned vehicle based on the driving environment and the number of follower vehicles, and can judge the relative position of the follower vehicle in the formation, thereby determining the final driving formation. The formation refers to the driving formation that the unmanned vehicle pre-arranges according to the driving environment and the number of follower vehicles before driving.

[0078] In an optional embodiment of the present invention, based on the driving environment and the number of the slave vehicles, the formation of the unmanned vehicle platoon is set using the following matrix:

[0079] ;

[0080] Among them, N represents the formation matrix of the unmanned vehicle team. represents the position of the first unmanned vehicle, represents the position of the second unmanned vehicle, represents the position of the mth unmanned vehicle, d represents the offset of the unmanned vehicle relative to the previous vehicle, and m represents the index number of the unmanned vehicle. Represents the position vectors of all unmanned vehicles.

[0081] Furthermore, the embodiments of the present invention can ensure that each vehicle in the unmanned vehicle can maintain a predetermined formation by locating the relative position of the slave vehicle in the formation, so as to maintain a safe distance between vehicles and reduce the risk of rear-end collision accidents. The relative position refers to the position of each slave vehicle relative to the main vehicle or other slave vehicles, such as the straight-line distance between the slave vehicle and the main vehicle or the front vehicle.

[0082] Optionally, the relative position positioning of the slave vehicle in the formation can be achieved by using a GPS positioning system.

[0083] The embodiment of the present invention determines the driving formation of the unmanned vehicle based on the navigation trajectory data and the relative position, thereby reducing the air resistance of the unmanned vehicle during driving and improving driving efficiency and vehicle safety. The driving formation refers to the way in which the vehicles in the unmanned vehicle are arranged in a specific geometric shape or sequence, such as a column formation, an inverted triangle formation, a square formation, etc.

[0084] As an embodiment of the present invention, determining the driving formation of the unmanned vehicle based on the navigation trajectory data and the relative position includes: identifying the driving trajectory of the unmanned vehicle based on the navigation trajectory data; dispatching the driving road condition information and formation formation of the unmanned vehicle according to the driving trajectory; adaptively adjusting the formation formation based on the driving road condition information and the relative position to obtain an adaptive formation; and determining the driving formation of the unmanned vehicle in combination with the driving road condition information and the adaptive formation.

[0085] Among them, the driving trajectory refers to the actual driving path of each vehicle in the unmanned vehicle, the driving road condition information refers to the collection of various real-time road conditions and environmental factors that affect the driving of the unmanned vehicle, such as road conditions, traffic flow, road obstacles, etc., and the adaptive formation refers to the vehicles in the unmanned vehicle automatically adjusting their own formation to adapt to the current environment according to the real-time driving road condition information and relative positions.

[0086] Optionally, based on the navigation trajectory data, the driving trajectory identification of the unmanned vehicle can be achieved using a path planning algorithm, and according to the driving trajectory, the driving road condition information scheduling of the unmanned vehicle can be obtained through a high-precision map, and according to the driving road condition information and the relative position, the adaptive adjustment of the formation formation can be achieved using a model predictive control algorithm.

[0087] The formation recovery module 103 is used to identify the task vehicle in the driving formation and the formation position of the task vehicle, build a side vehicle trajectory prediction network of the task vehicle based on the formation position, and set the formation recovery rules of the unmanned vehicle according to the side vehicle trajectory prediction network and the formation position.

[0088] The embodiment of the present invention can identify the task vehicles in the driving formation and the formation positions of the task vehicles, so that the task vehicles can enter the formation according to the predetermined positions after completing the tasks, thereby maintaining the stability of the formation. The task vehicles refer to the unmanned vehicles in the formation that are assigned to perform designated tasks, and the formation positions refer to the specific positions of each unmanned vehicle in the formation.

[0089] Optionally, identification of the task vehicle in the driving formation can be achieved using a vehicle identification code, and identification of the formation position of the task vehicle can be obtained through a V2X wireless communication system.

[0090] Furthermore, the embodiment of the present invention can better coordinate the driving actions of the slave vehicles by constructing a side vehicle trajectory prediction network of the task vehicle based on the formation position, so as to avoid affecting the normal driving of the side vehicles when the vehicle leaves the team or fails. The side vehicle trajectory prediction network refers to the use of sensors, data analysis, machine learning and other technologies to predict the possible driving paths and behaviors of neighboring vehicles in the future.

[0091] As an embodiment of the present invention, the construction of a side vehicle trajectory prediction network for the task vehicle based on the formation position includes: calculating the relative position of the side vehicle of the task vehicle based on the formation position; identifying a non-task vehicle at the relative position of the side vehicle; analyzing the characteristic behavior of the non-task vehicle relative to the task vehicle; extracting the driving behavior of the task vehicle, and analyzing the driving impact of the driving behavior on the non-task vehicle based on the characteristic behavior; and constructing a side vehicle trajectory prediction network for the task vehicle based on the characteristic behavior and the driving impact.

[0092] Among them, the relative position of the adjacent vehicle refers to the orientation and position of the non-mission vehicle relative to the mission vehicle; the non-mission vehicle refers to the vehicle other than the mission vehicle in the unmanned vehicle formation; the characteristic behavior refers to the specific behavior pattern exhibited by the vehicle during driving, such as acceleration, steering tendency, lane keeping ability, etc.; the driving behavior refers to the actions taken by the driver of the mission vehicle under specific circumstances, such as changing lanes, accelerating, decelerating, avoiding obstacles, etc.; the driving impact refers to the impact of the driving behavior of the mission vehicle on the driving status of the surrounding non-mission vehicles, such as forcing other vehicles to slow down, change lanes or take emergency braking, etc.

[0093] Optionally, the analysis of the impact of the driving behavior on the non-task vehicle based on the characteristic behavior can be implemented using a deep learning model.

[0094] In an optional embodiment of the present invention, based on the formation position, the relative position of the task vehicle and the adjacent vehicle is calculated using the following formula:

[0095] ;

[0096] in, represents the relative position vector of the mission vehicle and the adjacent vehicle, represents the position coordinates of the sidecar j relative to the task vehicle, Indicates the position coordinates of the main vehicle relative to the formation position.

[0097] In another optional embodiment of the present invention, the characteristic behavior of the non-mission vehicle relative to the mission vehicle is analyzed using the following formula:

[0098] ;

[0099] Where T represents the characteristic behavior of the non-mission vehicle relative to the mission vehicle, represents the distance difference between the i-th non-mission vehicle and the mission vehicle, represents the speed difference between the i-th non-mission vehicle and the mission vehicle, represents the distance difference between the i-th non-mission vehicle and the mission vehicle acted on by function p, represents the speed difference between the i-th non-mission vehicle and the mission vehicle as function g, represents the total number of non-mission vehicles, and i represents the sum index of non-mission vehicles.

[0100] The embodiment of the present invention sets the formation recovery rules of the unmanned vehicles according to the trajectory prediction network of the adjacent vehicles and the formation positions, thereby ensuring that the unmanned vehicles can adapt to the changes in the driving environment at any time and maintain the formation during driving. The formation recovery rules refer to the measures taken to adjust and restore the formation when the formation vehicles deviate from the predetermined formation due to external interference or internal changes during the driving of the unmanned vehicle formation.

[0101] As an embodiment of the present invention, the formation recovery rules of the unmanned vehicle are set according to the side vehicle trajectory prediction network and the formation position, including: real-time monitoring of the formation vehicles of the unmanned vehicle, and identifying the vehicles leaving the formation; analyzing the impact of the vehicles leaving the formation on the formation of the unmanned vehicle according to the side vehicle trajectory prediction network; scheduling the communication connection data between the unmanned vehicle and the vehicles leaving the formation, and identifying the reason for the vehicles leaving the formation based on the communication connection data; setting the vehicle compensation mechanism of the vehicles leaving the formation according to the formation impact and the reason for leaving the formation; identifying the driving trajectory of the unmanned vehicle, and setting the path for the vehicles leaving the formation to return to the formation based on the driving trajectory and the vehicle compensation mechanism; and setting the formation recovery rules of the unmanned vehicle in combination with the formation position and the path for returning to the formation.

[0102] Among them, the formation vehicle refers to an unmanned vehicle group arranged in a certain formation for a specific task or trip, the leaving vehicle refers to a vehicle that deviates from the predetermined formation or temporarily leaves the formation due to various reasons during the formation driving process, the formation impact refers to the change of the driving path of the remaining vehicles, the adjustment of the formation density and the impact on the surrounding traffic caused by the leaving vehicle, the communication connection data refers to the data exchanged between unmanned vehicles through vehicle-to-vehicle communication technology, including information such as the vehicle's position, speed, driving status and intention, the leaving reason refers to the specific reason that caused the vehicle to leave the team, such as technical failure, emergency avoidance, special task requirements, etc., the vehicle compensation mechanism refers to the adjustment strategy taken for the impact of the leaving vehicle on the formation, the driving trajectory refers to the actual path of the unmanned vehicle during driving, including changes in position, speed and direction over time, and the re-entry path refers to the predetermined path for the leaving vehicle to rejoin the formation.

[0103] Optionally, real-time monitoring of the unmanned vehicle formation vehicles can be achieved using on-board sensors, the analysis of the impact of the leaving vehicle on the unmanned vehicle formation based on the side vehicle trajectory prediction network can be obtained through a risk assessment model, and the setting of the vehicle compensation mechanism for the leaving vehicle based on the formation impact and the leaving reason can be achieved using a communication coordination method.

[0104] The distress perception module 104 is used to perceive the driving state of the unmanned vehicle in real time in combination with the driving formation and the formation recovery rule, analyze the environmental change trend of the driving environment based on the navigation trajectory data, and set the distress perception mode of the unmanned vehicle in combination with the vehicle driving state and the environmental change trend.

[0105] The embodiment of the present invention combines the driving formation and the formation recovery rules to perceive the vehicle driving status of the unmanned vehicle in real time, and can obtain the real-time operation status of the mission vehicle and the non-mission vehicle, and timely discover potential safety problems. When a vehicle leaves the team, the real-time status information helps to quickly identify the vehicle that has left the team and activate the formation recovery rules to minimize the impact on the team. The vehicle driving status refers to the dynamic behavior and physical state of the formation vehicles during driving, such as driving speed, steering angle, fault code and warning, etc.

[0106] Optionally, in combination with the driving formation and the formation recovery rule, real-time perception of the driving state of the unmanned vehicle can be achieved using on-board sensors, such as a laser radar.

[0107] Furthermore, the embodiments of the present invention can enable the unmanned vehicle to predict and avoid potential safety risks and improve safety by analyzing the environmental change trend of the driving environment based on the navigation trajectory data. The environmental change trend refers to the changing laws and development trends of various relevant factors and conditions of the unmanned vehicle's driving environment within a certain time and space range, such as changes in traffic flow, weather changes, traffic rules changes, etc.

[0108] As an embodiment of the present invention, the analysis of the environmental change trend of the driving environment based on the navigation trajectory data includes: performing attribute classification on the navigation trajectory data to obtain classified data; extracting key features of the classified data, and collecting historical data of the driving environment based on the key features; identifying traffic data and weather data in the historical data; analyzing the traffic flow and vehicle behavior patterns of the driving environment based on the traffic data; analyzing the visibility level of the driving environment based on the weather data; and analyzing the environmental change trend of the driving environment in combination with the traffic flow, the vehicle behavior patterns and the visibility level.

[0109] Among them, the classified data refers to the labeled data obtained after classifying the navigation trajectory data according to certain attributes or characteristics, such as meteorological data, traffic information, etc. The key feature refers to the attributes or parameters extracted from the classified data that are crucial to understanding the driving environment, such as vehicle driving speed, air humidity, traffic flow, etc. The historical data refers to the data records about the driving environment collected in the past period of time, the traffic data refers to the data related to the traffic conditions identified from the historical data, such as vehicle flow, speed, accident records, road construction information, etc. The weather data refers to the data related to the weather conditions identified from the historical data, such as temperature, precipitation, wind speed, visibility, etc. The traffic flow refers to the number of vehicles on a specific time and road section, the vehicle behavior pattern refers to the behavior rules of the vehicle during driving, such as acceleration and deceleration habits, lane change frequency, parking behavior, etc. The visibility level refers to the maximum distance at which the driver or sensor can clearly see the object in front under specific weather and environmental conditions.

[0110] Optionally, the key feature extraction of the classified data can be achieved using principal component analysis, the traffic flow and vehicle behavior pattern analysis of the driving environment based on the traffic data can be achieved using regression analysis, and the visibility level analysis of the driving environment based on the weather data can be obtained through a prediction model, such as an ARIMA model.

[0111] The embodiment of the present invention sets the distress perception mode of the unmanned vehicle by combining the vehicle driving status and the environmental change trend, thereby predicting potential dangerous situations, adjusting the driving strategies of the convoy vehicles in a timely manner, and enhancing the robustness of the unmanned vehicle system when facing complex and unpredictable environments. The distress perception mode refers to a function or mode that can identify and respond to adverse conditions or emergency situations that affect the safe driving of the unmanned vehicle.

[0112] As an embodiment of the present invention, the distress perception mode of the unmanned vehicle is set in combination with the vehicle driving state and the environmental change trend, including: identifying the surrounding vehicles of the unmanned vehicle based on the vehicle driving state; analyzing the interaction relationship between the surrounding vehicles and the unmanned vehicle; identifying the potential risk factors of the unmanned vehicle according to the environmental change trend and the interaction relationship, and identifying the risk level of the potential risk factors; setting the risk response mode of the unmanned vehicle according to the vehicle driving state and the risk level; setting the distress perception mode of the unmanned vehicle in combination with the risk level and the risk response method.

[0113] Among them, the surrounding vehicles refer to all other vehicles near the driving path of the unmanned vehicle or in the surrounding area; the interactive relationship refers to the dynamic relationship between the unmanned vehicle and the surrounding vehicles, including relative position, speed, driving direction and possible interactions; the potential risk factors refer to various situations or conditions that may cause driving safety problems of the unmanned vehicle, such as sudden lane changes, emergency braking, traffic accidents, etc. of other vehicles; the risk level refers to the risk classification obtained after a quantitative assessment of the risk based on the possibility and severity of the potential risk factors; the risk response method refers to the strategy or action taken by the unmanned vehicle to deal with different risk levels, including deceleration, lane change, parking, detour, etc.

[0114] Optionally, the analysis of the interaction relationship between the surrounding vehicles and the unmanned vehicle can be implemented using traffic simulation software, such as VISSIM simulation software; the identification of potential risk factors of the unmanned vehicle based on the environmental change trend and the interaction relationship can be obtained through a risk assessment model, such as an event tree analysis model; and the setting of the risk response mode of the unmanned vehicle based on the vehicle driving status and the risk level can be implemented using a vehicle control and decision-making system, such as an Autoware vehicle control system.

[0115] The single-vehicle escape module 105 is used to identify the abnormal condition of the unmanned vehicle based on the distress perception mode; and set the single-vehicle escape mechanism of the unmanned vehicle according to the abnormal condition of the vehicle and the adjacent vehicle trajectory prediction network.

[0116] The embodiment of the present invention can ensure that each vehicle in the unmanned vehicle can operate normally by identifying the vehicle abnormality based on the distress perception mode, which helps to maintain the continuity and stability of the unmanned vehicle formation. The vehicle abnormality refers to a situation in which a certain vehicle in the unmanned vehicle is inconsistent with the normal driving state or preset behavior.

[0117] As an embodiment of the present invention, the identifying of vehicle abnormalities of the unmanned vehicle based on the distress perception mode includes: based on the distress perception mode, collecting historical driving data of the unmanned vehicle; identifying normal driving parameters of the unmanned vehicle based on the historical driving data; identifying the current driving state of the unmanned vehicle, and identifying the current behavior deviation of the unmanned vehicle in combination with the current driving state and the historical driving data; based on the normal driving parameters, setting an abnormal parameter threshold of the unmanned vehicle; and identifying vehicle abnormalities of the unmanned vehicle based on the abnormal parameter threshold and the current behavior deviation.

[0118] Among them, the historical driving data refers to the data collected during the past driving process of the following vehicle, the normal driving parameters refer to the parameter values ​​generated under the normal driving state of the unmanned vehicle determined based on the historical driving data, the current driving state refers to the actual driving condition of the vehicle at the current moment, the current behavior deviation refers to the difference between the current driving state of the vehicle and the normal driving parameters, and the abnormal parameter threshold refers to the limit value set based on the normal driving parameters.

[0119] Optionally, the current driving state identification of the unmanned vehicle can be achieved using a time series analysis method. By combining the current driving state and the historical driving data, the current behavior deviation identification of the unmanned vehicle can be obtained through a deviation calculation formula, such as the difference between the current driving parameters and the normal driving parameters of the historical data.

[0120] Furthermore, the embodiment of the present invention sets up a single-vehicle escape mechanism for the unmanned vehicle according to the abnormal condition of the vehicle and the adjacent vehicle trajectory prediction network, thereby ensuring that the abnormal vehicle can safely escape from the predicament, improving the adaptability of the unmanned vehicle system in the face of emergencies, and ensuring the stability and reliability of the entire traffic flow. The single-vehicle escape mechanism refers to the function that a single vehicle can automatically or under remote control escape from the predicament when an unmanned vehicle formation encounters an abnormal condition or predicament during driving.

[0121] As an embodiment of the present invention, the single-vehicle escape mechanism of the unmanned vehicle is set according to the vehicle abnormal condition and the adjacent vehicle trajectory prediction network, including: locating the abnormal single vehicle of the unmanned vehicle according to the vehicle abnormal condition; analyzing the predicament type of the abnormal single vehicle, and setting a predicament response layer for the abnormal single vehicle based on the predicament type; identifying the adjacent vehicle driving trajectory of the abnormal single vehicle according to the adjacent vehicle trajectory prediction network; setting an escape path for the abnormal single vehicle based on the adjacent vehicle driving trajectory; and setting the single-vehicle escape mechanism of the unmanned vehicle in combination with the predicament response layer and the escape path.

[0122] Among them, the abnormal single vehicle refers to a single vehicle that exhibits abnormal behavior or status in an unmanned vehicle flow, such as deviation from the scheduled route, abnormal driving parameters, system failure, etc. The dilemma type refers to the type of specific dilemma or problem encountered by the abnormal single vehicle. The dilemma response layer refers to an escape execution layer set for the abnormal single vehicle for different types of dilemmas. The driving trajectory of the adjacent vehicle refers to the predicted driving path of other vehicles around the abnormal single vehicle. The escape path refers to a path planned for the abnormal single vehicle to safely move from the current position to an obstacle-free area.

[0123] Optionally, based on the abnormal condition of the vehicle, the abnormal single vehicle positioning of the unmanned vehicle can be obtained through an abnormality detection algorithm, and the dilemma type analysis of the abnormal single vehicle can be implemented using a classification algorithm using specific behavioral characteristics of the abnormal single vehicle, such as a cluster analysis algorithm, and based on the dilemma type, the dilemma strain layer setting of the abnormal single vehicle can be obtained through the motor torque and tire pressure of the vehicle.

[0124] The automatic formation module 106 is used to perform automatic formation processing on the unmanned vehicles according to the predicament perception mode, the single-vehicle escape mechanism and the formation recovery rule to obtain an automatic formation result.

[0125] The embodiment of the present invention automatically forms a team of unmanned vehicles according to the predicament perception mode, the single-vehicle escape mechanism and the formation recovery rule to obtain an automatic formation result. This can effectively cope with external interference, maintain the stability of the unmanned vehicle team formation, and dynamically adjust the formation structure and control strategy to adapt to environmental changes, thereby enhancing the robustness of the unmanned vehicle system in the face of complex and changing environments. At the same time, through effective predicament perception and single-vehicle escape, the unmanned vehicle formation can reduce unnecessary stops and decelerations caused by avoiding obstacles or handling emergencies, thereby ensuring the continuous execution of tasks and reducing task delays. The automatic formation result refers to the state reached by the unmanned vehicle system after automatically adjusting the relative positions and driving formations between vehicles according to the predicament perception mode and the single-vehicle escape mechanism.

[0126] As an embodiment of the present invention, the unmanned vehicles are automatically formed into a team according to the distress perception pattern, the single-vehicle escape mechanism and the formation recovery rule to obtain an automatic formation result, including: identifying the vehicle driving distress of the unmanned vehicle according to the distress perception pattern; setting the triggering conditions of the single-vehicle escape mechanism and the risk avoidance method of the unmanned vehicle based on the vehicle driving distress; performing an escape process on the unmanned vehicle according to the triggering conditions and the risk avoidance method to obtain an escape result; performing automatic formation processing on the unmanned vehicle based on the escape result and the formation recovery rule to obtain an automatic formation result.

[0127] Among them, the vehicle driving difficulty refers to any situation that may affect the normal driving of the unmanned vehicle during driving; the trigger condition refers to the specific parameters or situations that are set to trigger the implementation of the single-vehicle escape mechanism, such as the sensor detecting an obstacle ahead; the avoidance method refers to the specific action taken by the unmanned vehicle to avoid danger or difficulties after identifying the trigger condition; the escape result refers to the state of the unmanned vehicle after executing the avoidance method.

[0128] Optionally, based on the vehicle's driving difficulties, the trigger condition setting of the single-vehicle escape mechanism can be implemented by utilizing a perception system alarm, and according to the trigger condition and the risk avoidance method, the escape process of the unmanned vehicle can be determined by the operator's remote intervention and control of the vehicle.

[0129] The present invention collects the navigation trajectory data of the master vehicle, identifies the driving environment and the number of slave vehicles of the unmanned vehicle according to the navigation trajectory data, and can clarify the roles of the master vehicle and the slave vehicles, thereby realizing efficient driving of the unmanned vehicle formation; secondly, the embodiment of the present invention sets the formation of the unmanned vehicle based on the driving environment and the number of slave vehicles, and can judge the relative position of the slave vehicle in the formation, thereby determining the final driving formation, reducing the air resistance of the unmanned vehicle during driving, and improving driving efficiency and vehicle safety; thirdly, the embodiment of the present invention sets the formation recovery rules of the unmanned vehicle according to the side vehicle trajectory prediction network and the formation position, so as to ensure that the unmanned vehicle can adapt to the changes in the driving environment at any time during driving and maintain the formation formation; further, the embodiment of the present invention perceives the vehicle driving state of the unmanned vehicle in real time by combining the driving formation and the formation recovery rules, and can obtain the real-time operation status of the task vehicle and the non-task vehicle, timely discover potential safety problems, and improve safety. At the same time, combined with the vehicle driving state and the environmental change trend, the dilemma perception mode of the unmanned vehicle is set to predict potential dangerous situations and adjust the formation vehicles in time. The driving strategy can enhance the robustness of the unmanned vehicle system in the face of complex and unpredictable environments; the embodiment of the present invention can identify the abnormal situation of the unmanned vehicle based on the distress perception mode, and set the single-vehicle escape mechanism of the unmanned vehicle according to the abnormal situation of the vehicle and the trajectory prediction network of the adjacent vehicle, so as to ensure that each vehicle in the unmanned vehicle can operate normally, which is helpful to maintain the continuity and stability of the unmanned vehicle formation, and ensure that the abnormal vehicle can safely escape from the distress, improve the adaptability of the unmanned vehicle system in the face of emergencies, and ensure the stability and reliability of the entire traffic flow; finally, the embodiment of the present invention can effectively respond to external interference, maintain the stability of the unmanned vehicle formation, and dynamically adjust the formation structure and control strategy to adapt to environmental changes. The robustness of the unmanned vehicle system in the face of complex and changeable environments is enhanced. At the same time, through effective distress perception and single-vehicle escape, the unmanned vehicle formation can reduce unnecessary parking and deceleration caused by avoiding obstacles or handling emergencies, which can ensure the continuous execution of tasks and reduce task delays. Therefore, the unmanned vehicle automatic platooning system and method provided by the embodiments of the present invention can improve the unmanned vehicle's ability to escape from trouble on its own.

[0130] Reference Figure 2FIG. 1 is a flow chart of an automatic platooning method for unmanned vehicles provided in an embodiment of the present invention. In this embodiment, the automatic platooning method for unmanned vehicles includes:

[0131] Acquire unmanned vehicles to be automatically formed into a team, determine the master vehicle and the slave vehicles of the unmanned vehicle, collect navigation trajectory data of the master vehicle, and identify the driving environment of the unmanned vehicle and the number of the slave vehicles according to the navigation trajectory data;

[0132] Based on the driving environment and the number of the slave vehicles, the formation of the unmanned vehicle is set, the relative position of the slave vehicle in the formation is located, and based on the navigation trajectory data and the relative position, the driving formation of the unmanned vehicle is determined;

[0133] Identify the task vehicle in the driving formation and the formation position of the task vehicle, build a side vehicle trajectory prediction network of the task vehicle based on the formation position, and set the formation recovery rule of the unmanned vehicle according to the side vehicle trajectory prediction network and the formation position;

[0134] In combination with the driving formation and the formation recovery rule, the driving state of the unmanned vehicle is perceived in real time, based on the navigation trajectory data, the environmental change trend of the driving environment is analyzed, and in combination with the vehicle driving state and the environmental change trend, the distress perception mode of the unmanned vehicle is set;

[0135] Based on the distress perception mode, identifying the abnormal condition of the unmanned vehicle; and setting a single vehicle escape mechanism of the unmanned vehicle according to the abnormal condition of the vehicle and the adjacent vehicle trajectory prediction network;

[0136] According to the distress perception mode, the single-vehicle escape mechanism and the formation recovery rule, the unmanned vehicles are automatically formed into a team to obtain an automatic formation result.

[0137] In the several embodiments provided by the present invention, it should be understood that the provided system and method can be implemented in other ways. For example, the system embodiment described above is only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0138] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. An unmanned vehicle automatic platooning system, characterized in that: The unmanned vehicle automatic formation system includes: a data acquisition module, a driving formation determination module, a formation recovery module, a predicament perception module, a single vehicle escape module, and an automatic formation module; The data acquisition module is used to acquire the unmanned vehicles to be automatically formed, determine the master vehicle and the slave vehicles of the unmanned vehicles, collect the navigation trajectory data of the master vehicle, and identify the driving environment of the unmanned vehicle and the number of the slave vehicles according to the navigation trajectory data; The driving formation determination module is used to set the formation of the unmanned vehicle based on the driving environment and the number of the slave vehicles, locate the relative position of the slave vehicle in the formation, and determine the driving formation of the unmanned vehicle based on the navigation trajectory data and the relative position; The formation recovery module is used to identify the task vehicle in the driving formation and the formation position of the task vehicle, build a side vehicle trajectory prediction network of the task vehicle based on the formation position, and set the formation recovery rules of the unmanned vehicle according to the side vehicle trajectory prediction network and the formation position; The distress perception module is used to perceive the driving state of the unmanned vehicle in real time in combination with the driving formation and the formation recovery rule, analyze the environmental change trend of the driving environment based on the navigation trajectory data, and set the distress perception mode of the unmanned vehicle in combination with the vehicle driving state and the environmental change trend; The single vehicle escape module is used to identify the abnormal condition of the unmanned vehicle based on the distress perception mode; and set the single vehicle escape mechanism of the unmanned vehicle according to the abnormal condition of the vehicle and the adjacent vehicle trajectory prediction network; The automatic formation module is used to automatically form a formation for the unmanned vehicles according to the distress perception mode, the single-vehicle escape mechanism and the formation recovery rule to obtain an automatic formation result.

2. The unmanned vehicle automatic platooning system according to claim 1, characterized in that: The determining the driving formation of the unmanned vehicle based on the navigation trajectory data and the relative position includes: Based on the navigation trajectory data, identifying the driving trajectory of the unmanned vehicle; According to the driving trajectory, dispatching the driving road condition information and formation of the unmanned vehicle; Based on the driving road condition information and the relative position, adaptively adjusting the formation to obtain an adaptive formation; The driving formation of the unmanned vehicle is determined by combining the driving road condition information and the adaptive formation.

3. The unmanned vehicle automatic platooning system according to claim 1, characterized in that: The step of constructing a vehicle trajectory prediction network for the task vehicle based on the formation position includes: Based on the formation position, calculating the relative position of the task vehicle and the adjacent vehicles; Identifying a non-mission vehicle at a relative position of the adjacent vehicle; analyzing characteristic behaviors of the non-mission vehicle relative to the mission vehicle; Extracting the driving behavior of the task vehicle, and analyzing the driving impact of the driving behavior on the non-task vehicle based on the characteristic behavior; According to the characteristic behavior and the driving impact, a vehicle trajectory prediction network of the task vehicle is constructed.

4. The unmanned vehicle automatic platooning system according to claim 1, characterized in that: The step of setting the formation recovery rules of the unmanned vehicle according to the adjacent vehicle trajectory prediction network and the formation position includes: Monitoring the vehicles in the convoy of the unmanned vehicles in real time and identifying the vehicles that have left the convoy; Analyzing the impact of the departing vehicle on the formation of the unmanned vehicle according to the adjacent vehicle trajectory prediction network; Dispatching communication connection data between the unmanned vehicle and the leaving vehicle, and identifying the leaving reason of the leaving vehicle based on the communication connection data; According to the formation impact and the reason for leaving the team, a vehicle compensation mechanism is set for the leaving vehicle; Identify the driving trajectory of the unmanned vehicle, and set a path for the vehicle leaving the team to return to the team based on the driving trajectory and the vehicle compensation mechanism; In combination with the formation position and the re-entry path, a formation recovery rule for the unmanned vehicle is set.

5. The unmanned vehicle automatic platooning system according to claim 1, characterized in that: The analyzing the environmental change trend of the driving environment based on the navigation trajectory data includes: Attribute classification is performed on the navigation trajectory data to obtain classification data; Extracting key features of the classified data, and collecting historical data of the driving environment based on the key features; identifying traffic data and weather data in the historical data; Analyzing the traffic flow and vehicle behavior patterns of the driving environment based on the traffic data; analyzing a visibility level of the driving environment based on the weather data; The environmental change trend of the driving environment is analyzed in combination with the traffic flow, the vehicle behavior pattern and the visibility level.

6. The unmanned vehicle automatic platooning system according to claim 1, characterized in that: The step of setting the distress perception mode of the unmanned vehicle in combination with the vehicle driving state and the environmental change trend includes: Based on the driving state of the vehicle, identifying vehicles in the surrounding environment of the unmanned vehicle; Analyzing the interaction relationship between the surrounding environment vehicles and the unmanned vehicle; Identifying potential risk factors of the unmanned vehicle according to the environmental change trend and the interactive relationship, and identifying risk levels of the potential risk factors; Setting a risk response mode of the unmanned vehicle according to the vehicle driving state and the risk level; In combination with the risk level and the risk response method, a distress perception mode of the unmanned vehicle is set.

7. The unmanned vehicle automatic platooning system according to claim 1, characterized in that: The identifying the abnormal situation of the unmanned vehicle based on the distress perception mode includes: Based on the dilemma perception mode, collecting historical driving data of the unmanned vehicle; Identifying normal driving parameters of the unmanned vehicle based on the historical driving data; Identifying a current driving state of the unmanned vehicle, and identifying a current behavior deviation of the unmanned vehicle by combining the current driving state with the historical driving data; Based on the normal driving parameters, setting an abnormal parameter threshold of the unmanned vehicle; According to the abnormal parameter threshold and the current behavior deviation, an abnormal condition of the unmanned vehicle is identified.

8. The unmanned vehicle automatic platooning system according to claim 1, characterized in that: The step of setting a single vehicle escape mechanism of the unmanned vehicle according to the abnormal condition of the vehicle and the adjacent vehicle trajectory prediction network includes: According to the abnormal condition of the vehicle, locate the abnormal single vehicle of the unmanned vehicle; Analyze the dilemma type of the abnormal bicycle, and set a dilemma response layer for the abnormal bicycle based on the dilemma type; According to the side vehicle trajectory prediction network, identifying the side vehicle driving trajectory of the abnormal single vehicle; Based on the driving trajectory of the adjacent vehicle, setting an escape path for the abnormal single vehicle; In combination with the distress response layer and the escape path, a single-vehicle escape mechanism for the unmanned vehicle is set.

9. The unmanned vehicle automatic platooning system according to claim 1, characterized in that: The automatic formation processing of the unmanned vehicles according to the distress perception mode, the single vehicle escape mechanism and the formation recovery rule to obtain the automatic formation result includes: identifying a driving difficulty of the unmanned vehicle according to the difficulty perception mode; Based on the vehicle driving predicament, setting the triggering conditions of the single-vehicle escape mechanism and setting the risk avoidance method of the unmanned vehicle; According to the trigger condition and the risk avoidance method, the unmanned vehicle is rescued to obtain a rescue result; Based on the escape result and the formation recovery rule, the unmanned vehicles are automatically formed into a team to obtain an automatic formation result.

10. An automatic platooning method for unmanned vehicles, characterized in that: The method comprises: Acquire unmanned vehicles to be automatically formed into a team, determine the master vehicle and the slave vehicles of the unmanned vehicle, collect navigation trajectory data of the master vehicle, and identify the driving environment of the unmanned vehicle and the number of the slave vehicles according to the navigation trajectory data; Based on the driving environment and the number of the slave vehicles, the formation of the unmanned vehicle is set, the relative position of the slave vehicle in the formation is located, and based on the navigation trajectory data and the relative position, the driving formation of the unmanned vehicle is determined; Identify the task vehicle in the driving formation and the formation position of the task vehicle, build a side vehicle trajectory prediction network of the task vehicle based on the formation position, and set the formation recovery rule of the unmanned vehicle according to the side vehicle trajectory prediction network and the formation position; In combination with the driving formation and the formation recovery rule, the driving state of the unmanned vehicle is perceived in real time, based on the navigation trajectory data, the environmental change trend of the driving environment is analyzed, and in combination with the vehicle driving state and the environmental change trend, the distress perception mode of the unmanned vehicle is set; Based on the distress perception mode, identifying the abnormal condition of the unmanned vehicle; and setting a single vehicle escape mechanism of the unmanned vehicle according to the abnormal condition of the vehicle and the adjacent vehicle trajectory prediction network; According to the distress perception mode, the single-vehicle escape mechanism and the formation recovery rule, the unmanned vehicles are automatically formed into a team to obtain an automatic formation result.

Images