Automatic driving vehicle formation control method and system based on working condition identification

Abstract

The invention discloses an automatic driving vehicle formation control method and system based on working condition identification, and relates to the technical field of automatic driving. Deploying a sensor; the leading vehicle is used for sensing the surrounding environment; judging the driving condition of the current motorcade; obtaining a relative position relationship between adjacent vehicles in the motorcade; setting a time headway to obtain an expected distance between adjacent vehicles; arbitrating the expected distance; calculating a distance error between adjacent vehicles; whether the speed of the rear vehicle needs to be adjusted is judged, and if yes, the speed of the rear vehicle is adjusted; and if not, continuously driving according to the current vehicle speed. Complete automatic driving modification is carried out on a small part of vehicles in a motorcade, the current environment is perceived through the leading vehicle, the perceived information is issued to all following vehicles in the motorcade in a broadcast mode, the following vehicles in the motorcade are assisted to carry out planning control, and a motorcade control task can be completed.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a method and system for controlling an autonomous driving vehicle formation based on operating condition recognition. Background Art

[0002] With the continuous maturity of automotive electronic technology and more thoughts from all walks of life on the safety, energy and environment of truck road transportation, more and more truck manufacturers are entering the field of autonomous driving, trying to seize the initiative in terms of truck intelligence, safety and energy saving.

[0003] Vehicle platooning is a new vehicle driving technology that can greatly improve traffic safety, effectively solve traffic congestion and improve traffic efficiency, reduce fleet energy consumption, reduce driver labor intensity, and improve social productivity. However, in the process of hardware modification for autonomous driving vehicle platooning, there is a large demand for high-precision sensors such as lidar, millimeter-wave radar, and cameras, which are expensive. Communication equipment needs to be dedicated and high-bandwidth, which is expensive. Computing hardware needs to be high-performance and may require redundant design, which is costly. It also involves other hardware costs such as brake system upgrades and electrical system modifications, resulting in a high cost for hardware modification for autonomous driving vehicle platooning. At the same time, professional installation and debugging and calibration also consume high labor costs and time resources, which to a certain extent limits its large-scale promotion and application.

[0004] Although autonomous vehicle platooning technology already exists in the existing technology, it is highly dependent on field and satellite communications. Since the coverage of field communications is often limited, field signals may be interrupted or unstable in some remote areas or places with weak communication infrastructure, making it impossible for truck platoons to obtain accurate road conditions and instructions in a timely manner, thus affecting the coordinated operation of the entire platoon; although satellite communications can provide coverage in a wider area, they also have some inherent defects. For example, signals are easily interfered by weather conditions. For example, heavy rain, thick clouds, etc. may cause signal attenuation or even interruption. In addition, the bandwidth of satellite communications is limited, and congestion may occur when processing large amounts of data at the same time, resulting in delayed information transmission.

[0005] In addition, existing autonomous vehicle platooning technology has little application of real-time operating condition recognition technology, and real-time operating condition recognition technology itself is still in the development stage. It is difficult to accurately identify complex and changing road environments, weather conditions, and traffic flows. For example, under different lighting conditions, the sensor's perception of the surrounding environment may be affected, resulting in deviations in the judgment of real-time operating conditions. At the same time, different road conditions such as curves, ramps, bridges, etc. will also bring challenges to operating condition recognition, because the vehicle dynamics characteristics in these places will change, and existing technologies may not be able to capture these changes in a timely and accurate manner. Summary of the invention

[0006] The present invention provides an automatic driving vehicle formation control method and system based on working condition recognition, so as to overcome at least one technical problem existing in the prior art.

[0007] In one aspect, the present invention provides a method for controlling a formation of autonomous driving vehicles based on operating condition recognition, comprising:

[0008] Setting a vehicle formation, determining a leading vehicle and a following vehicle located behind the leading vehicle;

[0009] Deploying sensors on the leading vehicle and the following vehicle respectively, so that the leading vehicle has a full autonomous driving function and the following vehicle has a partial autonomous driving function;

[0010] Using the leading vehicle to sense the surrounding environment, and outputting the sensed result to the formation local area network;

[0011] Sending the driving data of the leading vehicle and the following vehicle and the warning information during the driving process to the formation local area network;

[0012] Using the perception results, the driving data and the warning information, determining the driving condition of the current fleet;

[0013] Obtain the relative position relationship d between adjacent vehicles in the fleet actual ;

[0014] Assume the headway time is t h , then the expected distance d between adjacent vehicles desired Represented as d desired =v 后 ·t h , where v 后 Indicates the actual speed of the following vehicle;

[0015] According to the current driving conditions of the fleet, the threshold d of the expected distance is set to d1≤d≤d2;

[0016] According to the threshold d of the expected distance, the expected distance d desired Conduct arbitration and obtain the expected distance d after arbitration desired1 ;

[0017] According to the expected distance d after arbitration desired1 and relative position relationship d actual , calculate the distance error between adjacent vehicles as e d =d desired -d actual ;

[0018] Based on the distance error and a predetermined safe distance, determine whether it is necessary to adjust the speed of the following vehicle. If so, adjust the speed of the following vehicle; if not, continue to drive at the current speed.

[0019] Optionally, the ratio of the leading vehicle to the following vehicle is 1:4.

[0020] Optionally, use the perception result, the driving data, and the warning information to determine the driving condition of the current vehicle fleet. Specifically:

[0021] According to the formula Calculate the acceleration state c of the vehicle and the average vehicle speed within a predetermined time period where v i represents the speed of vehicle i, represents the average value of the absolute values of the accelerations of all vehicles within a predetermined time period, represents the average value of the absolute values of the jerk of all vehicles within a predetermined time period;

[0022] Set a speed threshold V and an acceleration state threshold C;

[0023] Compare the average vehicle speed with the speed threshold V and the acceleration state c with the acceleration state threshold C respectively to obtain the driving condition of the current vehicle fleet.

[0024] Optionally, set the speed threshold as v1 ≤ V ≤ v2, and the acceleration state threshold as c1 ≤ C ≤ c2;

[0025] When and c ≤ c1, it indicates that the current vehicle fleet is in a low-speed smooth condition; the threshold d of the desired distance is 10 ≤ d ≤ 20;

[0026] When and c ≤ c1, it indicates that the current vehicle fleet is in a medium-speed smooth condition; the threshold d of the desired distance is 20 ≤ d ≤ 30;

[0027] When and c ≤ c1, it indicates that the current vehicle fleet is in a high-speed smooth condition; the threshold d of the desired distance is 30 ≤ d ≤ 50;

[0028] When and c1 < c, it indicates that the current vehicle fleet is in a low-speed complex condition; the threshold d of the desired distance is 20 ≤ d ≤ 30;

[0029] When and c1 < c, it indicates that the current vehicle fleet is in a medium-speed complex condition; the threshold d of the desired distance is 30 ≤ d ≤ 40;

[0030] When When c1 < c, it indicates that the current vehicle fleet is in a high-speed complex working condition; the threshold d of the desired distance is 50 ≤ d ≤ 60.

[0031] Optionally, obtain the relative position relationship d between adjacent vehicles in the vehicle fleet actual , specifically:

[0032] Determine whether there is a high-precision positioning GNSS signal. If so, directly use the high-precision positioning GNSS signal to obtain the relative position relationship d between adjacent vehicles in the vehicle fleet actual ; if not, use the sensors deployed on the vehicle to collect and calculate the relative position relationship d between adjacent vehicles in the vehicle fleet actual .

[0033] Optionally, use the sensors deployed on the vehicle to collect and calculate the relative position relationship d between adjacent vehicles in the vehicle fleet actual , specifically:

[0034] Use the sensors deployed on the leading vehicle to collect the positions of each following vehicle in the vehicle fleet;

[0035] Use the sensors deployed on each following vehicle to collect the distance from the vehicle to the vehicle in front and the offset distance of the vehicle relative to the lane line;

[0036] Perform data fusion on the positions of each following vehicle in the vehicle fleet, the distance from the vehicle to the vehicle in front, and the offset distance of the vehicle relative to the lane line to obtain the relative position relationship d between adjacent vehicles in the vehicle fleet actual .

[0037] Optionally, the desired distance d after arbitration desired1 is:

[0038] d desired1 =(d desired < d2)? ((d desired < d1)? d1 : d desired ): d2.

[0039] Optionally, determine whether it is necessary to adjust the speed of the following vehicle according to the distance error and the predetermined safety distance, specifically:

[0040] Judge whether the distance error is greater than the predetermined safety distance. If so, it means that it is not necessary to adjust the speed of the following vehicle; if not, it means that it is necessary to adjust the speed of the following vehicle.

[0041] Optionally, adjust the speed of the following vehicle, specifically:

[0042] According to the speed v of the following vehicle 后 and the speed v of the vehicle in front 前, calculate the speed error e v =v 前 -v 后 ;

[0043] According to the speed error, based on the proportional-integral-derivative control algorithm Adjust the acceleration of the rear vehicle; where K p Represents the proportionality coefficient, K i Indicates the integral coefficient, K d represents the differential coefficient.

[0044] On the other hand, the present invention also provides an automatic driving vehicle formation control system based on working condition recognition, comprising:

[0045] A first setting module is used to set the formation of vehicles and determine a leading vehicle and a following vehicle located behind the leading vehicle;

[0046] A deployment module, used to deploy sensors on the leading vehicle and the following vehicle respectively, so that the leading vehicle has a complete automatic driving function and the following vehicle has a partial automatic driving function;

[0047] A perception module, used to use the leading vehicle to perceive the surrounding environment and output the perception result to the formation local area network;

[0048] A sending module, used for sending the driving data of the leading vehicle and the following vehicles and the warning information during the driving process to the formation local area network;

[0049] A first judgment module, used to judge the driving condition of the current fleet by using the perception result, the driving data and the warning information;

[0050] The acquisition module is used to obtain the relative position relationship d between adjacent vehicles in the fleet. actual ;

[0051] Setting module, used to set the headway time as t h , then the expected distance d between adjacent vehicles desired Represented as d desired =v 后 ·t h , where v 后 Indicates the actual speed of the following vehicle;

[0052] The second setting module sets the threshold d of the expected distance to d1≤d≤d2 according to the current driving conditions of the fleet;

[0053] The arbitration module is used to determine the expected distance d according to the threshold d of the expected distance. desiredConduct arbitration and obtain the expected distance d after arbitration desired1 ;

[0054] A calculation module is used to calculate the expected distance d after the arbitration. desired1 and relative position relationship d actual , calculate the distance error between adjacent vehicles as e d =d desired -d actual ;

[0055] The second judgment module is used to judge whether it is necessary to adjust the speed of the following vehicle according to the distance error and the predetermined safety distance. If so, the speed of the following vehicle is adjusted; if not, the vehicle continues to travel at the current speed.

[0056] The innovative features of the embodiments of the present invention include:

[0057] 1. In this embodiment, a small number of vehicles in the fleet are fully modified for automatic driving, and the leading vehicle senses the current environment and broadcasts the sensed information to all following vehicles in the formation, thereby assisting the following vehicles in the formation to perform planning and control, thereby completing the fleet control task. This can avoid the cost of excessive redundant installation of sensors, which is not only conducive to reducing costs, but also reduces the pressure on communication bandwidth within the fleet, which is one of the innovative points of the embodiment of the present invention.

[0058] 2. In this embodiment, the working condition identification method based on pattern recognition judges the overall operating status of the current fleet, and combines the relative position relationship of each vehicle in the fleet to assist the vehicle in switching driving modes, thereby realizing the overall coordinated control of the fleet to ensure the overall safety and economy of the vehicle, which is one of the innovative points of the embodiment of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 A flow chart of a formation control method provided by an embodiment of the present invention;

[0061] Figure 2 The specific arrangement form of the vehicle formation provided by the embodiment of the present invention;

[0062] Figure 3 A flowchart of arbitration provided by an embodiment of the present invention;

[0063] Figure 4 A flow chart for determining the current working condition provided by an embodiment of the present invention;

[0064] Figure 5 A flow chart for calculating relative position relationship provided by an embodiment of the present invention;

[0065] Figure 6 A flow chart of adjusting the speed of a rear vehicle provided by an embodiment of the present invention;

[0066] Figure 7 A schematic structural diagram of a formation control system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0067] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0068] It should be noted that the terms "including" and "having" and any variations thereof in the embodiments of the present invention and the accompanying drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0069] The embodiment of the present invention discloses a method and system for controlling a formation of autonomous driving vehicles based on working condition recognition, which are described in detail below.

[0070] Figure 1 For a flow chart of a formation control method provided by an embodiment of the present invention, please refer to Figure 1 , the automatic driving vehicle formation control method based on working condition recognition provided by an embodiment of the present invention includes:

[0071] Step 1: Set the vehicle formation and determine the leading vehicle and the following vehicles behind the leading vehicle;

[0072] Step 2: Deploy sensors on the leading vehicle and the following vehicle respectively, so that the leading vehicle has a full autonomous driving function and the following vehicle has a partial autonomous driving function;

[0073] Step 3: Use the leading vehicle to perceive the surrounding environment and output the perception results to the formation local area network;

[0074] Step 4: Send the driving data of the leading vehicle and the following vehicles and the warning information during the driving process to the formation local area network;

[0075] Step 5: Use the perception results, driving data and warning information to determine the current driving conditions of the fleet;

[0076] Step 6: Obtain the relative position relationship d between adjacent vehicles in the fleet actual ;

[0077] Step 7: Let the headway time be t h , then the expected distance d between adjacent vehicles desired Represented as d desired =v 后 ·t h , where v 后 Indicates the actual speed of the following vehicle;

[0078] Step 8: According to the current driving conditions of the fleet, the threshold d of the expected distance is set to d1≤d≤d2;

[0079] Step 9: According to the threshold d of the expected distance, the expected distance d desired Conduct arbitration and obtain the expected distance d after arbitration desired1 ;

[0080] Step 10: According to the expected distance d after arbitration desired1 and relative position relationship d actual , calculate the distance error between adjacent vehicles as e d =d desired -d actual ;

[0081] Step 11: Based on the distance error and the predetermined safety distance, determine whether the speed of the following vehicle needs to be adjusted. If so, adjust the speed of the following vehicle; if not, continue driving at the current speed.

[0082] Specifically, please refer to Figure 1 , the automatic driving vehicle formation control method based on working condition recognition provided by the embodiment of the present invention first forms a truck formation with network-connected intelligent driving function through step 1, and determines the leading vehicle and the following vehicle. In the driving direction of the convoy, the following vehicle is located behind the leading vehicle. The specific arrangement of the vehicle formation can be referred to Figure 2 .

[0083] It should be noted that in a truck formation, there is a leading vehicle and multiple following vehicles, that is, the ratio of the leading vehicle to the following vehicles is 1:n. When the value of n is too small, more leading vehicles need to be set up, so more vehicles need to be fully modified for autonomous driving, which is not conducive to reducing the modification cost of the autonomous driving truck formation; when the value of n is too large, a leading vehicle needs to control more following vehicles, and for the following vehicles far away from the leading vehicle, there may be problems such as unstable signal transmission leading to control errors.

[0084] In summary, n needs to be set to an appropriate value according to the actual working conditions of the road. In this embodiment, for example, the ratio of the leading vehicle to the following vehicles can be set to 1:4. In this way, by using the method of leading the vehicle, only a small number of vehicles in the fleet need to be modified for relatively complete autonomous driving, and these modified vehicles can serve as the leading vehicles, responsible for sensing the external environment, identifying the current working conditions, and issuing upper-level control instructions, so as to lead the fleet, which can effectively reduce the modification cost of the autonomous driving vehicle formation.

[0085] After determining the leading vehicle and the following vehicle, sensors can be deployed on the leading vehicle and the following vehicle respectively through step 2, so that the leading vehicle and the following vehicle have the automatic driving function. In the same formation, the following vehicle follows the leading vehicle, that is, the leading vehicle needs to perform the automatic driving function, while the following vehicle only needs to perform the following task. Therefore, when modifying the leading vehicle and the following vehicle, the leading vehicle should have a complete automatic driving function, while the following vehicle only needs to have a partial automatic driving function. In order to enable the leading vehicle to have a complete automatic driving function, the leading vehicle needs to be equipped with cameras (such as dual front-view cameras, rear-view cameras, blind spot cameras), laser radars, blind spot laser radars, millimeter-wave radars, ultrasonic radars, combined navigation and other sensors, as well as intelligent driving domain controllers, and equipped with V2X communication equipment. For other following vehicles, they only need to be equipped with basic sensors such as cameras and millimeter-wave radars and V2X communication equipment.

[0086] After the modification is completed, in step 3, the leading vehicle is used to perceive the surrounding environment, and the perception results are output to the formation local area network. Here, the perception results include, for example, target recognition, target posture recognition, target motion recognition, road information recognition, traffic sign recognition, drivable area recognition, obstacle recognition warning, etc. When transmitting the perception results, they are transmitted in the form of a dictionary. Table 1 is an example of a perception result output format.

[0087] Table 1

[0088] type Target Name property Property Value pedestrian 1 Pose, size, speed, ... […],[…],… vehicle 1 Pose, size, speed, ... […],[…],… … … … … obstacle 1 Pose, size,... […],[…],… ... ... ... ...

[0089] In order to identify the driving conditions of the vehicles, the sensors installed on the vehicles will collect real-time driving data of the vehicles during driving. In step 4, the collected driving data of the leading vehicle and the following vehicle will be sent to the formation local area network. At the same time, the warning information during driving will be sent to the formation local area network through the V2X communication equipment.

[0090] After the formation local area network receives the sensing results, driving data and warning information, it can use the sensing results, driving data and warning information in step 5 to determine the working condition of the current fleet, so as to adjust the driving parameters of the current fleet according to different working conditions. However, it should be noted that when adjusting the fleet, in addition to considering the driving conditions, the relative position relationship between adjacent vehicles in the fleet also needs to be considered. Therefore, the present invention also obtains the relative position relationship between adjacent vehicles in the fleet through step 6.

[0091] After obtaining the relative position relationship between the working condition and the adjacent vehicles, in step 7, it is assumed that the expected headway is t h , then according to the actual speed of the following vehicle, the expected distance d between adjacent vehicles can be calculated desired is d desired =v 后 ·t h , where v 后 Indicates the actual speed of the following vehicle.

[0092] After obtaining the expected distance, it is necessary to determine whether the expected distance is reasonable through the threshold. Therefore, the threshold d of the expected distance is set in step 8. It should be noted that the threshold d of the expected distance is related to the current driving conditions of the fleet. Therefore, it is necessary to set the threshold d of the expected distance to d1≤d≤d2 according to the current driving conditions of the fleet obtained in step 5. Then in step 9, the threshold d of the expected distance is set to d1≤d≤d2. desired1 =(d desired <d2)?((d desired <d1)?d1:d desired ):d2 compares the expected distance d desired The threshold d of the expected distance is used to achieve the expected distance d desired Arbitration, the expected distance after arbitration is d desired1 .

[0093] like Figure 3 For a flowchart of arbitration provided by an embodiment of the present invention, please refer to Figure 3 , first determine the expected distance d desired Is it less than the threshold d2? If not, the threshold d2 is used as the expected distance d after arbitration. desired1 ; If it is less than the threshold d2, then the expected distance d is determined desiredIs it less than the threshold d1? If it is less than the threshold d1, then the threshold d1 is used as the expected distance d after arbitration. desired1 ; If it is not less than the threshold d1, then the expected distance d desired As the expected distance after arbitration d desired1 .

[0094] Get the expected distance d after arbitration desired1 and relative position relationship d actual Then, in step 10, according to the expected distance d after arbitration desired1 and relative position relationship d actual , the distance error between adjacent vehicles is calculated as e d =d desired -d actual . Then in step 11, by comparing the distance error with the predetermined safety distance, it can be determined whether the speed of the following vehicle needs to be adjusted. For example, by comparing the distance error with the predetermined safety distance, when the distance error is greater than the predetermined safety distance, it means that the current distance between adjacent vehicles meets the expected distance. Therefore, there is no need to adjust the speed of the following vehicle, and the vehicle continues to travel at the current speed. When the distance error is less than or equal to the predetermined safety distance, it means that the current distance between adjacent vehicles does not meet the expected distance. Therefore, it is necessary to adjust the speed of the following vehicle to expand or shorten the following distance as needed, so as to meet different driving conditions and enable the convoy to travel safely and stably.

[0095] The automatic driving vehicle formation control method based on working condition recognition provided by the present invention can complete the task of fleet control by completely modifying a small number of vehicles in the fleet for automatic driving, sensing the current environment through the leading vehicle, and broadcasting the sensed information to all the following vehicles in the formation, thereby assisting the following vehicles in the formation to perform planning and control, and thus avoiding the excessive redundant installation costs of sensors, which is not only conducive to reducing costs, but also reduces the pressure on the communication bandwidth within the fleet. In addition, the working condition recognition method based on pattern recognition of the present invention judges the overall operation status of the current fleet, and assists the vehicle in switching the driving mode in combination with the relative position relationship of each vehicle in the fleet, thereby realizing the coordinated control of the entire fleet to ensure the overall safety and economy of the vehicle.

[0096] Optionally, Figure 4 For a flow chart of determining the current working condition provided by an embodiment of the present invention, please refer to Figure 1 and Figure 4 In step 5, the driving condition of the current fleet is determined by using the perception results, driving data and warning information. Specifically, step 51, according to the formula Calculate the vehicle's acceleration state c and the average vehicle speed within a predetermined period of time Among them, v i represents the speed of vehicle i, represents the average value of the absolute values of the accelerations of all vehicles within a predetermined time period, represents the average value of the absolute values of the jerk of all vehicles within a predetermined time period; Step 52, set the speed threshold V and the acceleration state threshold C; Step 53, respectively compare the average vehicle speed and the speed threshold V, and the acceleration state c and the acceleration state threshold C, to obtain the driving condition of the current vehicle fleet.

[0097] Preferably, set the speed threshold as v1 ≤ V ≤ v2, and the acceleration state threshold as c1 ≤ C ≤ c2; when and c ≤ c1, it indicates that the current vehicle fleet is in a low-speed smooth driving condition; the threshold d of the desired distance is 10 ≤ d ≤ 20; when and c ≤ c1, it indicates that the current vehicle fleet is in a medium-speed smooth driving condition; the threshold d of the desired distance is 20 ≤ d ≤ 30; when and c ≤ c1, it indicates that the current vehicle fleet is in a high-speed smooth driving condition; the threshold d of the desired distance is 30 ≤ d ≤ 50; when and c1 < c, it indicates that the current vehicle fleet is in a low-speed complex driving condition; the threshold d of the desired distance is 20 ≤ d ≤ 30; when and c1 < c, it indicates that the current vehicle fleet is in a medium-speed complex driving condition; the threshold d of the desired distance is 30 ≤ d ≤ 40; when and c1 < c, it indicates that the current vehicle fleet is in a high-speed complex driving condition; the threshold d of the desired distance is 50 ≤ d ≤ 60.

[0098] Specifically, please refer to Figure 4 , when judging the driving condition of the vehicle fleet, first calculate the acceleration state of the vehicle and the average vehicle speed within a predetermined time period through Step 51. In this embodiment, according to the formula calculate the acceleration state c of the vehicle and the average vehicle speed within a predetermined time period; among them, v i represents the speed of vehicle i, represents the average value of the absolute values of the accelerations of all vehicles within a predetermined time period, represents the average value of the absolute values of the jerk of all vehicles within a predetermined time period.

[0099] After obtaining the above parameters, judge the driving condition of the vehicle fleet by comparing each parameter with its threshold. Therefore, in this embodiment, through Step 52, set the speed threshold V and the acceleration state threshold C. Then in Step 53, respectively compare the average vehicle speed and the speed threshold V, and the acceleration state c and the acceleration state threshold C, to obtain the driving condition of the current vehicle fleet.

[0100] Assume that the speed threshold is v1 ≤ V ≤ v2, and the acceleration state threshold is c1 ≤ C ≤ c2. Regarding the driving conditions of the vehicle platoon, in this embodiment, it is set that when and c ≤ c1, it indicates that the current vehicle platoon is in a low-speed smooth condition; the threshold d of the expected distance is set as 10 ≤ d ≤ 20; when and c ≤ c1, it indicates that the current vehicle platoon is in a medium-speed smooth condition; the threshold d of the expected distance is 20 ≤ d ≤ 30; when and c ≤ c1, it indicates that the current vehicle platoon is in a high-speed smooth condition; the threshold d of the expected distance is 30 ≤ d ≤ 50; when and c1 < c, it indicates that the current vehicle platoon is in a low-speed complex condition; the threshold d of the expected distance is 20 ≤ d ≤ 30; when and c1 < c, it indicates that the current vehicle platoon is in a medium-speed complex condition; the threshold d of the expected distance is 30 ≤ d ≤ 40; when and c1 < c, it indicates that the current vehicle platoon is in a high-speed complex condition; the threshold d of the expected distance is 50 ≤ d ≤ 60.

[0101] Optionally, please refer to Figure 1 , in step 6, obtain the relative position relationship between adjacent vehicles in the vehicle platoon. Specifically: determine whether there is a high-precision positioning GNSS signal. If so, directly use the high-precision positioning GNSS signal to obtain the relative position relationship between adjacent vehicles in the vehicle platoon; if not, use the sensors deployed on the vehicles to collect and calculate the relative position relationship between adjacent vehicles in the vehicle platoon.

[0102] Specifically, usually, the relative position relationship between adjacent vehicles in the vehicle platoon is obtained by using the high-precision positioning GNSS signal. However, when entering working conditions such as tunnels and caves, there may be a situation where there is no high-precision positioning GNSS signal. Based on this situation, when obtaining the relative position relationship between adjacent vehicles in the vehicle platoon in this embodiment, first determine whether there is a high-precision positioning GNSS signal. If so, directly use the high-precision positioning GNSS signal to obtain the relative position relationship between adjacent vehicles in the vehicle platoon. When there is no high-precision positioning GNSS signal, use the sensors deployed on the vehicles to collect the position information of each vehicle, and calculate the relative position relationship between adjacent vehicles in the vehicle platoon according to the position information. In this way, even in an environment with poor satellite communication, the relative positions of vehicles within the vehicle platoon can be autonomously determined through the multi-vehicle information fusion method, realizing the collaborative control of the entire vehicle platoon and ensuring the stability of the vehicle platoon during driving.

[0103] Optionally, Figure 5 is a flowchart for calculating the relative position relationship provided by an embodiment of the present invention. Please refer to Figure 5, using sensors deployed on the vehicles to collect and calculate the relative position relationship between adjacent vehicles in the convoy, specifically: step 61, using the sensors deployed on the leading vehicle to collect the position of each following vehicle in the convoy; step 62, using the sensors deployed on each following vehicle to collect the distance from the vehicle to the leading vehicle and the offset distance of the vehicle relative to the lane line; step 63, performing data fusion on the position of each following vehicle in the convoy, the distance from the vehicle to the leading vehicle and the offset distance of the vehicle relative to the lane line to obtain the relative position relationship between adjacent vehicles in the convoy.

[0104] Specifically, please refer to Figure 5 When there is no high-precision positioning GNSS signal, it is necessary to use the sensors deployed on the vehicle to collect and calculate the relative position relationship between each adjacent vehicle in the convoy. In step 61, the sensor deployed on the leading vehicle, such as a laser radar, roughly scans the position of each following vehicle in the convoy and uploads it to the local area network. Then in step 62, the distance from the vehicle to the leading vehicle and the offset distance of the vehicle relative to the lane line are collected using the sensors deployed on each following vehicle, and uploaded to the local area network.

[0105] After obtaining the above information, in step 63, by integrating the position of each following vehicle in the convoy, the distance from the vehicle to the front vehicle, and the offset distance of the vehicle relative to the lane line, the relative position relationship between each adjacent vehicle in the convoy can be obtained.

[0106] When performing data fusion, first set the state vector x k =[d k ,v rel,k ] T , where d k represents the actual distance between the two vehicles at time k, v rel , k Represents the relative speed between the front vehicle and the vehicle. Perform prior state estimation Δt represents the time step. The distance at the current moment is predicted based on the distance at the previous moment plus the product of the relative speed and the time step. When predicting, the relative speed remains unchanged.

[0107] The prior error covariance estimate is defined as P k - =FP k-1 F T +Q, where the state transfer matrix Q is the process noise covariance matrix, which can be set according to actual conditions. The observation vector is defined as where d measured,k is the distance between the two vehicles measured by radar, v own,k is the speed of the vehicle at time k, v front,kis the speed of the preceding vehicle at time k, which can be estimated by the rate of change of distance measured by the radar. The observation equation is defined as z k =Hx k +v k , where the observation matrix v k is the observation noise, whose covariance matrix is ​​R.

[0108] In the update step, the gain is calculated as: K k =P k - H T (HP k - H T +R) -1 ; The posterior state estimate is: The posterior error covariance estimate is: P k =(IK k H)P k - , where I represents the identity matrix.

[0109] Optionally, Figure 6 For a flow chart of adjusting the rear vehicle speed provided by an embodiment of the present invention, please refer to Figure 1 and Figure 6 In step 11, the speed of the rear vehicle is adjusted, specifically: step 011, according to the speed v of the rear vehicle 后 and the speed of the preceding vehicle v 前 , calculate the speed error e v =v 前 -v 后 ; Step 012, according to the speed error, based on the proportional integral differential control algorithm Adjust the acceleration of the rear vehicle; where K p Represents the proportionality coefficient, K i Indicates the integral coefficient, K d represents the differential coefficient.

[0110] Specifically, please refer to Figure 6 When the speed of the following vehicle needs to be adjusted, first in step 011, according to the speed v of the following vehicle 后 and the speed of the preceding vehicle v 前 , calculate the speed error e v =v 前 -v 后 . Get the speed error e v Then, in step 012, according to the speed error e v , adjust the acceleration of the rear vehicle. When adjusting the acceleration, you can use the proportional integral derivative control algorithm To adjust the acceleration of the following vehicle.p Represents the proportionality coefficient, K i Indicates the integral coefficient, K d Represents the differential coefficient. During the adjustment process, different control effects can be achieved by adjusting the above coefficients.

[0111] Based on the same inventive concept, the present invention also provides an automatic driving vehicle formation control system based on working condition recognition. Figure 7 For a structural diagram of a formation control system provided by an embodiment of the present invention, please refer to Figure 7 , the automatic driving vehicle formation control system 100 based on working condition recognition provided by an embodiment of the present invention includes:

[0112] The first setting module 101 is used to set the formation of vehicles and determine the leading vehicle and the following vehicles behind the leading vehicle;

[0113] A deployment module 102, for deploying sensors on the leading vehicle and the following vehicle respectively, so that the leading vehicle has a full autonomous driving function and the following vehicle has a partial autonomous driving function;

[0114] A perception module 103 is used to use the leading vehicle to perceive the surrounding environment and output the perception result to the formation local area network;

[0115] The sending module 104 is used to send the driving data of the leading vehicle and the following vehicles and the warning information during the driving process to the formation local area network;

[0116] The first judgment module 105 is used to judge the driving condition of the current fleet by using the sensing result, driving data and warning information;

[0117] An acquisition module 106 is used to acquire the relative position relationship between adjacent vehicles in the fleet;

[0118] Setting module 107, used to set the headway time as t h , then the expected distance d between adjacent vehicles desired Represented as d desired =v 后 ·t h , where v 后 Indicates the actual speed of the following vehicle;

[0119] The second setting module 108 sets the threshold d of the expected distance to d1≤d≤d2 according to the current driving condition of the fleet;

[0120] The arbitration module 109 is used to determine the expected distance d according to the threshold d of the expected distance. desired Conduct arbitration and obtain the expected distance d after arbitration desired1 ;

[0121] The calculation module 110 is used to calculate the expected distance d after arbitration. desired1 and relative position relationship, the distance error between adjacent vehicles is calculated as e d =d desired -d actual ;

[0122] The second judgment module 111 is used to judge whether the speed of the following vehicle needs to be adjusted according to the distance error and the predetermined safety distance. If so, the speed of the following vehicle is adjusted; if not, the vehicle continues to travel at the current speed.

[0123] Specifically, please refer to Figure 7 , the automatic driving vehicle formation control system 100 based on working condition recognition provided by the embodiment of the present invention first forms a truck formation with network-connected intelligent driving function through the first setting module 101, and determines the leading vehicle and the following vehicle. In the driving direction of the convoy, the following vehicle is located behind the leading vehicle. The specific arrangement of the vehicle formation can be referred to Figure 2 .

[0124] It should be noted that in a truck formation, there is a leading vehicle and multiple following vehicles, that is, the ratio of the leading vehicle to the following vehicles is 1:n. When the value of n is too small, more leading vehicles need to be set up, so more vehicles need to be fully modified for autonomous driving, which is not conducive to reducing the modification cost of the autonomous driving truck formation; when the value of n is too large, a leading vehicle needs to control more following vehicles, and for the following vehicles far away from the leading vehicle, there may be problems such as unstable signal transmission leading to control errors.

[0125] In summary, n needs to be set to an appropriate value according to the actual working conditions of the road. In this embodiment, for example, the ratio of the leading vehicle to the following vehicles can be set to 1:4. In this way, by using the method of leading the vehicle, only a small number of vehicles in the fleet need to be modified for relatively complete autonomous driving, and these modified vehicles can serve as the leading vehicles, responsible for sensing the external environment, identifying the current working conditions, and issuing upper-level control instructions, so as to lead the fleet, which can effectively reduce the modification cost of the autonomous driving vehicle formation.

[0126] After determining the leading vehicle and the following vehicle, the sensors can be deployed on the leading vehicle and the following vehicle respectively through the deployment module 102, so that the leading vehicle and the following vehicle have the automatic driving function. In the same formation, the following vehicle follows the leading vehicle, that is, the leading vehicle needs to perform the automatic driving function, and the following vehicle only needs to perform the following task. Therefore, when modifying the leading vehicle and the following vehicle, the leading vehicle should have a complete automatic driving function, while the following vehicle only needs to have a partial automatic driving function. In order to make the leading vehicle have a complete automatic driving function, the leading vehicle needs to be installed with cameras (such as dual front-view cameras, rear-view cameras, blind-filling cameras), laser radars, blind-filling laser radars, millimeter-wave radars, ultrasonic radars, combined navigation and other sensors, as well as intelligent driving domain controllers, and equipped with V2X communication equipment. For other following vehicles, it is only necessary to equip them with basic sensors such as cameras, millimeter-wave radars and V2X communication equipment.

[0127] After the modification is completed, the perception module 103 uses the leading vehicle to perceive the surrounding environment and outputs the perception results to the formation local area network. Here, the perception results include, for example, target recognition, target posture recognition, target motion recognition, road information recognition, traffic sign recognition, drivable area recognition, obstacle recognition warning, etc. When transmitting the perception results, they are transmitted in the form of a dictionary, such as Table 1 which is an example of a perception result output format.

[0128] In order to identify the driving conditions of the vehicles, the sensors installed on the vehicles will collect real-time driving data of the vehicles during driving. The collected driving data of the leading vehicle and the following vehicles will be sent to the formation local area network through the sending module 104. At the same time, the warning information during driving will be sent to the formation local area network through the V2X communication device.

[0129] After the formation local area network receives the sensing results, driving data and warning information, the first judgment module 105 uses the sensing results, driving data and warning information to determine the working condition of the current fleet, so that the driving parameters of the current fleet can be adjusted according to different working conditions. However, it should be noted that when adjusting the fleet, in addition to considering the driving conditions, the relative position relationship between adjacent vehicles in the fleet also needs to be considered. Therefore, the present invention also obtains the relative position relationship between adjacent vehicles in the fleet through the acquisition module 106.

[0130] After obtaining the relative position relationship between the working condition and the adjacent vehicles, the setting module 107 assumes that the expected headway is t h , then according to the actual speed of the following vehicle, the expected distance d between adjacent vehicles can be calculated desired is d desired =v 后 ·t h, where v 后 Indicates the actual speed of the following vehicle.

[0131] After obtaining the expected distance, it is necessary to determine whether the expected distance is reasonable through a threshold. Therefore, the threshold d of the expected distance is first set by the second setting module 108. It should be noted that the threshold d of the expected distance is related to the current driving conditions of the fleet. Therefore, it is necessary to set the threshold d of the expected distance to d1≤d≤d2 according to the current driving conditions of the fleet. Then the arbitration module 109 determines the expected distance through d desired1 =(d desired <d2)?((d desired <d1)?d1:d desired ):d2 compares the expected distance d desired The threshold d of the expected distance is used to achieve the expected distance d desired Arbitration, the expected distance after arbitration is d desired1 .

[0132] Please refer to Figure 3 , first determine the expected distance d desired Is it less than the threshold d2? If not, the threshold d2 is used as the expected distance d after arbitration. desired1 ; If it is less than the threshold d2, then the expected distance d is determined desired Is it less than the threshold d1? If it is less than the threshold d1, then the threshold d1 is used as the expected distance d after arbitration. desired1 ; If it is not less than the threshold d1, then the expected distance d desired As the expected distance after arbitration d desired1 .

[0133] Get the expected distance d after arbitration desired1 and relative position relationship d actual After that, the calculation module 110 calculates the expected distance d after arbitration. desired1 and relative position relationship d actual , the distance error between adjacent vehicles is calculated as e d =d desired -d actual. The second judgment module 111 can determine whether it is necessary to adjust the speed of the following vehicle by comparing the distance error with the predetermined safety distance. For example, by comparing the distance error with the predetermined safety distance, when the distance error is greater than the predetermined safety distance, it means that the current distance between adjacent vehicles meets the expected distance. Therefore, there is no need to adjust the speed of the following vehicle, and the vehicle continues to travel at the current speed. When the distance error is less than or equal to the predetermined safety distance, it means that the current distance between adjacent vehicles does not meet the expected distance. Therefore, it is necessary to adjust the speed of the following vehicle to expand or shorten the following distance as needed, so as to meet different driving conditions and enable the convoy to travel safely and stably.

[0134] The automatic driving vehicle formation control system based on working condition recognition provided by the present invention can complete the task of fleet control by completely modifying a small number of vehicles in the fleet for automatic driving, sensing the current environment through the leading vehicle, and broadcasting the sensed information to all the following vehicles in the formation, thereby assisting the following vehicles in the formation to perform planning and control, and thus avoiding the excessive redundant installation costs of sensors, which is not only conducive to reducing costs, but also reduces the pressure on the communication bandwidth within the fleet. In addition, the working condition recognition method based on pattern recognition of the present invention judges the overall operation status of the current fleet, and assists the vehicle in switching the driving mode in combination with the relative position relationship of each vehicle in the fleet, thereby realizing the coordinated control of the entire fleet to ensure the overall safety and economy of the vehicle.

[0135] Those skilled in the art can understand that the accompanying drawings are only schematic diagrams of an embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0136] Those skilled in the art can understand that the modules in the device in the embodiment can be distributed in the device in the embodiment according to the description of the embodiment, or can be changed accordingly and located in one or more devices different from the embodiment. The modules in the above embodiment can be combined into one module, or can be further divided into multiple sub-modules.

[0137] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for controlling a formation of autonomous vehicles based on working condition recognition, characterized in that: include: Setting a vehicle formation, determining a leading vehicle and a following vehicle located behind the leading vehicle; Deploying sensors on the leading vehicle and the following vehicle respectively, so that the leading vehicle has a full autonomous driving function and the following vehicle has a partial autonomous driving function; Using the leading vehicle to sense the surrounding environment, and outputting the sensed result to the formation local area network; Sending the driving data of the leading vehicle and the following vehicle and the warning information during the driving process to the formation local area network; Using the perception results, the driving data and the warning information, determining the driving condition of the current fleet; Obtain the relative position relationship d between adjacent vehicles in the fleet actual ; Assume the headway time is t h , then the expected distance d between adjacent vehicles desired Represented as d desired =v 后 ·t h , where v 后 Indicates the actual speed of the following vehicle; According to the current driving conditions of the fleet, the threshold d of the expected distance is set to d1≤d≤d2; According to the threshold d of the expected distance, the expected distance d desired Conduct arbitration and obtain the expected distance d after arbitration desired1 ; According to the expected distance d after arbitration desired1 and relative position relationship d actual , calculate the distance error between adjacent vehicles as e d =d desired -d actual ; According to the distance error and the predetermined safety distance, it is determined whether the speed of the following vehicle needs to be adjusted. If so, the speed of the following vehicle is adjusted; if not, the vehicle continues to travel at the current speed.

2. The method for controlling a formation of autonomous vehicles based on working condition recognition according to claim 1, characterized in that: The ratio of the leading vehicle to the following vehicles is 1:

4.

3. The method for controlling a formation of autonomous driving vehicles based on working condition recognition according to claim 2, characterized in that: The driving condition of the current fleet is determined by using the perception result, the driving data and the warning information, specifically: According to the formula Calculate the vehicle's acceleration state c and the average vehicle speed within a predetermined period of time Among them, v i represents the speed of vehicle i, It represents the average value of the absolute value of acceleration of all vehicles in a predetermined period of time. It represents the average value of the absolute values ​​of the acceleration of all vehicles in a predetermined period; Set the speed threshold V and the acceleration state threshold C; Compare the average speeds The speed threshold V, the acceleration state c and the acceleration state threshold C are used to obtain the driving condition of the current fleet.

4. The method for controlling a formation of autonomous vehicles based on working condition recognition according to claim 3, characterized in that: Set the velocity threshold to v1≤V≤v2 and the acceleration state threshold to c1≤C≤c2; when When c≤c1, it means that the current fleet is in a low-speed smooth condition; the threshold d of the expected distance is 10≤d≤20; when When c≤c1, it means that the current fleet is in a medium-speed smooth condition; the threshold d of the expected distance is 20≤d≤30; when When c≤c1, it means that the current fleet is in high-speed smooth operation; the threshold d of the expected distance is 30≤d≤50; When and c1 < c, it indicates that the current vehicle fleet is in a low-speed complex working condition; the threshold d of the desired distance is 20 ≤ d ≤ 30; When and c1 < c, it indicates that the current vehicle fleet is in a medium-speed complex working condition; the threshold value d of the expected distance is 30 ≤ d ≤ 40; When and c1 < c, it indicates that the current vehicle fleet is in a high-speed complex working condition; the threshold d of the expected distance is 50 ≤ d ≤ 60.

5. The method for controlling a formation of autonomous driving vehicles based on working condition recognition according to claim 1, characterized in that: Obtain the relative position relationship d between adjacent vehicles in the fleet actual , specifically: Determine whether there is a high-precision positioning GNSS signal. If so, directly use the high-precision positioning GNSS signal to obtain the relative position relationship between adjacent vehicles in the fleet. actual If not, the sensors deployed on the vehicles are used to collect and calculate the relative position relationship d between the adjacent vehicles in the convoy. actual .

6. The method for controlling a formation of autonomous vehicles based on working condition recognition according to claim 5, characterized in that: The sensors deployed on the vehicles are used to collect and calculate the relative position relationship between adjacent vehicles in the convoy. actual , specifically: Using sensors deployed on the leading vehicle, the positions of the following vehicles in the convoy are collected; The sensors deployed on each following vehicle are used to collect the distance between the vehicle and the preceding vehicle and the offset distance of the vehicle relative to the lane line. The data of the position of each following vehicle in the convoy, the distance from the vehicle to the preceding vehicle, and the offset distance of the vehicle relative to the lane line are fused to obtain the relative position relationship d between adjacent vehicles in the convoy. actual .

7. The method for controlling a formation of autonomous vehicles based on working condition recognition according to claim 1, characterized in that: Expected distance after arbitration d desired1 for: d desired1 =(d desired <d2)?((d desired <d1)?d1:d desired ):d2。 8. The method for controlling a formation of autonomous vehicles based on working condition recognition according to claim 1, characterized in that: According to the distance error and the predetermined safety distance, it is determined whether the speed of the following vehicle needs to be adjusted, specifically: It is determined whether the distance error is greater than the predetermined safety distance. If so, it indicates that the speed of the following vehicle does not need to be adjusted; if not, it indicates that the speed of the following vehicle needs to be adjusted.

9. The method for controlling a formation of autonomous driving vehicles based on working condition recognition according to claim 1, characterized in that: Adjust the speed of the rear vehicle, specifically: According to the speed v of the following vehicle 后 and the speed of the preceding vehicle v 前 , calculate the speed error e v =v 前 -v 后 ; According to the speed error, based on the proportional-integral-derivative control algorithm Adjust the acceleration of the rear vehicle; where K p Represents the proportionality coefficient, K i Indicates the integral coefficient, K d represents the differential coefficient.

10. An automatic driving vehicle formation control system based on working condition recognition, characterized in that: include: A first setting module is used to set the formation of vehicles and determine a leading vehicle and a following vehicle located behind the leading vehicle; A deployment module, used to deploy sensors on the leading vehicle and the following vehicle respectively, so that the leading vehicle has a complete automatic driving function and the following vehicle has a partial automatic driving function; A perception module, used to use the leading vehicle to perceive the surrounding environment and output the perception result to the formation local area network; A sending module, used for sending the driving data of the leading vehicle and the following vehicles and the warning information during the driving process to the formation local area network; A first judgment module, used to judge the driving condition of the current fleet by using the perception result, the driving data and the warning information; The acquisition module is used to obtain the relative position relationship d between adjacent vehicles in the fleet. actual ; Setting module, used to set the headway time as t h , then the expected distance d between adjacent vehicles desired Represented as d desired =v 后 ·t h , where v 后 Indicates the actual speed of the following vehicle; The second setting module sets the threshold d of the expected distance to d1≤d≤d2 according to the current driving conditions of the fleet; The arbitration module is used to determine the expected distance d according to the threshold d of the expected distance. desired Conduct arbitration and obtain the expected distance d after arbitration desired1 ; A calculation module is used to calculate the expected distance d after the arbitration. desired1 and relative position relationship d actual , calculate the distance error between adjacent vehicles as e d =d desired -d actual ; The second judgment module is used to judge whether it is necessary to adjust the speed of the following vehicle according to the distance error and the predetermined safety distance. If so, the speed of the following vehicle is adjusted; if not, the vehicle continues to travel at the current speed.

Images