Vehicle obstacle avoidance control method, device and vehicle

Abstract

Embodiments of the present invention provide a vehicle obstacle avoidance control method, device, and vehicle, including, after obtaining vehicle status information and road surface information during the current vehicle driving process, real-time calculation of the road adhesion coefficient based on the vehicle status information, construction of a driving risk field model based on the road adhesion coefficient and road surface information, calculation of an optimal reference path for obstacle avoidance based on the driving risk field model, and control of the vehicle to track the optimal reference path for obstacle avoidance. This application fully considers the impact of the road adhesion coefficient on vehicle obstacle avoidance control and, therefore, constructs a driving risk field model for the road adhesion coefficient. Vehicle obstacle avoidance control based on this driving risk field model can significantly improve obstacle avoidance accuracy, thereby effectively reducing the driver's driving risk.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of vehicle obstacle avoidance, and in particular to a vehicle obstacle avoidance control method, device, and vehicle. Background Art

[0002] According to research, about 90% of traffic accidents are related to drivers. Vehicle active obstacle avoidance technology can enable vehicles to autonomously avoid obstacles according to obstacle avoidance paths, thereby improving driving safety. At present, the main vehicle obstacle avoidance methods include risk field method (artificial potential field method), optimal control method and random search method. Among them, the risk field method is widely used in vehicle obstacle avoidance due to its advantages such as simple calculation, relatively smooth planned path and good real-time performance.

[0003] The driving environment of a vehicle is complex and ever-changing. Sudden changes in road adhesion conditions often occur during driving. The most typical example is when the weather changes from sunny to rainy, in which case the road adhesion conditions will change from high to low. In such scenarios, if the impact of road adhesion changes on the risk field is not considered, the accuracy of obstacle avoidance control will be seriously affected, thereby greatly increasing the driving risk for the driver. Summary of the Invention

[0004] In view of this, the embodiments of the present invention provide a vehicle obstacle avoidance control method, device and vehicle, which can improve the accuracy of vehicle obstacle avoidance, thereby reducing the driver's driving risk and allowing the driver to travel safely.

[0005] In a first aspect, an embodiment of the present invention provides a vehicle obstacle avoidance control method, wherein the method is applied to an on-board controller; the method includes:

[0006] Obtain vehicle status information and road surface information during the current vehicle driving process;

[0007] Real-time calculation of road adhesion coefficient based on vehicle status information;

[0008] Construct a driving risk field model based on the road adhesion coefficient and road surface information;

[0009] Calculate the optimal reference path for obstacle avoidance based on the driving risk field model;

[0010] Control the vehicle to follow the optimal reference path to avoid obstacles.

[0011] The above-mentioned step of calculating the road adhesion coefficient in real time based on vehicle status information includes:

[0012] Extract vehicle tire information and vehicle speed information from vehicle status information in real time;

[0013] Input the vehicle tire information and vehicle speed information into a pre-trained tire model, and output the tire adhesion coefficient corresponding to each tire of the vehicle through the tire model;

[0014] The road adhesion coefficient is obtained by averaging multiple tire adhesion coefficients.

[0015] The above steps of constructing a driving risk field model based on the road adhesion coefficient and road surface information include:

[0016] Extracting road boundary information, first position information of the vehicle heading towards the target position, current second position information of the obstacle, relative information between the obstacle and the vehicle, and current third position information of the vehicle from the road surface information;

[0017] Constructing a road boundary risk field model based on road boundary information and third position information;

[0018] constructing a target gravitational field model based on the first position information and the third position information;

[0019] Constructing an obstacle risk field model based on the road adhesion coefficient, the second position information, the relative information, and the third position information;

[0020] A driving risk field model is constructed based on the road boundary risk field model, target gravity field model and obstacle risk field model.

[0021] The above steps of constructing the driving risk field model based on the road boundary risk field model, the target gravity field model, and the obstacle risk field model include:

[0022] The road boundary risk field model, target gravity field model and obstacle risk field model are weightedly calculated to obtain the driving risk field model.

[0023] The above steps of calculating the optimal reference path for obstacle avoidance based on the driving risk field model include:

[0024] Perform negative gradient derivation on the driving risk field model to obtain multiple obstacle avoidance path points of the initial obstacle avoidance path;

[0025] Calculate the optimal reference path based on multiple obstacle avoidance path points.

[0026] The above step of calculating the optimal reference path based on multiple obstacle avoidance path points includes:

[0027] Perform multiple multinomial fitting calculations on multiple obstacle avoidance path points to obtain the optimal reference path.

[0028] In a second aspect, an embodiment of the present invention provides a vehicle obstacle avoidance control device, wherein the device is applied to a vehicle-mounted server; the device includes:

[0029] An acquisition module is used to obtain vehicle status information and road surface information during the current vehicle driving process;

[0030] A first calculation module is used to calculate the road adhesion coefficient in real time based on vehicle state information;

[0031] A construction module is used to construct a driving risk field model based on the road adhesion coefficient and road surface information;

[0032] The second calculation module is used to calculate the optimal reference path for obstacle avoidance based on the driving risk field model;

[0033] The control module is used to control the vehicle to track the optimal reference path for obstacle avoidance.

[0034] The first calculation module is further configured to: extract vehicle tire information and vehicle speed information from the vehicle status information in real time;

[0035] Input the vehicle tire information and vehicle speed information into a pre-trained tire model, and output the tire adhesion coefficient corresponding to each tire of the vehicle through the tire model;

[0036] The road adhesion coefficient is obtained by averaging multiple tire adhesion coefficients.

[0037] The above-mentioned construction module is further used to: extract road boundary information, first position information of the vehicle heading towards the target position, current second position information of the obstacle, relative information between the obstacle and the vehicle, and current third position information of the vehicle from the road surface information;

[0038] Constructing a road boundary risk field model based on road boundary information and third position information;

[0039] constructing a target gravitational field model based on the first position information and the third position information;

[0040] Constructing an obstacle risk field model based on the road adhesion coefficient, the second position information, the relative information, and the third position information;

[0041] A driving risk field model is constructed based on the road boundary risk field model, target gravity field model and obstacle risk field model.

[0042] In a third aspect, an embodiment of the present invention provides a vehicle, which is equipped with an on-board controller, and the on-board controller is used to execute the above-mentioned vehicle obstacle avoidance control method.

[0043] The embodiments of the present invention bring the following beneficial effects:

[0044] Embodiments of the present invention provide a vehicle obstacle avoidance control method, device, and vehicle. After obtaining vehicle status information and road surface information during the current vehicle driving process, the method calculates the road adhesion coefficient in real time based on the vehicle status information, constructs a driving risk field model based on the road adhesion coefficient and road surface information, and calculates an optimal reference path for obstacle avoidance based on the driving risk field model. The vehicle is then controlled to track the optimal reference path for obstacle avoidance. This application fully considers the impact of the road adhesion coefficient on vehicle obstacle avoidance control and, therefore, constructs a driving risk field model for the road adhesion coefficient. Vehicle obstacle avoidance control based on this driving risk field model can significantly improve obstacle avoidance accuracy, thereby effectively reducing the driver's driving risk. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A flow chart of a vehicle obstacle avoidance control method provided in this embodiment;

[0046] Figure 2 A flow chart of another vehicle obstacle avoidance control method provided in this embodiment;

[0047] Figure 3 A three-dimensional schematic diagram of a road boundary risk field model provided in this embodiment;

[0048] Figure 4 A three-dimensional schematic diagram of a target gravitational field model provided in this embodiment;

[0049] Figure 5 A three-dimensional schematic diagram of an obstacle risk field modeling provided in this embodiment;

[0050] Figure 6 A three-dimensional schematic diagram of a driving risk field model provided in this embodiment;

[0051] Figure 7 This is a schematic structural diagram of a vehicle obstacle avoidance control device provided in this embodiment. DETAILED DESCRIPTION

[0052] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings 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 making creative efforts shall fall within the scope of protection of the present invention.

[0053] To facilitate understanding of the embodiments of the present invention, specific embodiments will be further explained below with reference to the accompanying drawings. The embodiments do not limit the embodiments of the present invention.

[0054] This embodiment provides a vehicle obstacle avoidance control method, wherein the method is applied to a vehicle controller; see Figure 1 The flowchart of a vehicle obstacle avoidance control method shown in FIG. 1 includes the following steps:

[0055] S102, obtaining vehicle status information and road surface information during the current vehicle driving process;

[0056] In actual use, the onboard controller communicates with a sensor group on the vehicle. The sensor group is used to collect external data of the vehicle, namely road information, and detect dynamic data of the vehicle during driving, namely vehicle status information. The sensor group includes, but is not limited to, at least one of a camera, a lidar, a millimeter-wave radar, a GPS (Global Positioning System), an IMU (Inertial Measurement Unit), a velocity sensor, and an acceleration sensor.

[0057] The on-board controller is used to obtain data from the sensor group. All sensors in the sensor group transmit data at a high frequency while the vehicle is driving. The on-board controller is also used to wirelessly communicate with the cloud server and exchange various information.

[0058] The on-board controller is also used to calculate the road adhesion coefficient based on the data from the sensor group, perform path planning and decision-making based on the road adhesion coefficient, and generate vehicle control instructions based on the planned optimal reference path, thereby controlling the vehicle to drive safely and avoid obstacles.

[0059] S104, calculating the road adhesion coefficient in real time based on the vehicle state information;

[0060] The road adhesion coefficient is the ratio of adhesion to wheel normal pressure (perpendicular to the road surface). It can be regarded as the static friction coefficient between the tire and the road surface, which is roughly equivalent to the friction coefficient. Different road adhesion states have a significant impact on the vehicle's obstacle avoidance path and obstacle avoidance effect. At the same vehicle speed, the lower the road adhesion coefficient, the smaller the lateral acceleration during obstacle avoidance, the smaller the standard deviation of the lateral acceleration, and the smoother the obstacle avoidance effect.

[0061] S106, constructing a driving risk field model based on the road adhesion coefficient and road surface information;

[0062] Driving risk field models are an effective means of assessing vehicle driving safety. Currently, mainstream models fail to consider the impact of road adhesion coefficient on driving safety. This neglect of the impact of road adhesion coefficient on driving risk is clearly inconsistent with actual conditions and can seriously impact the safety of obstacle avoidance control. Therefore, during active vehicle safety control, it is necessary to calculate the road adhesion coefficient in real time and incorporate this calculated road adhesion coefficient into the driving risk field model. This effectively improves the model's accuracy, enabling precise vehicle control for safe obstacle avoidance and significantly reducing driver risk.

[0063] S108, calculating an optimal reference path for obstacle avoidance based on the driving risk field model;

[0064] The premise of vehicle obstacle avoidance is to control the vehicle to follow the optimal reference path calculated by the driving risk field model for safe driving. The optimal reference path can be understood as the obstacle avoidance path with the lowest driving risk.

[0065] S110, controlling the vehicle to track the optimal reference path to perform obstacle avoidance driving.

[0066] In this embodiment, the prediction model can be used to track the optimal reference path to control the front wheel angle of the vehicle for obstacle avoidance.

[0067] In specific implementation, the construction process of the prediction model is as follows: First, considering the lateral and longitudinal motions of the vehicle during obstacle avoidance as well as the yaw motion of the vehicle, ignoring the influence of the suspension and vertical motion, a three-degree-of-freedom vehicle model is established, as shown in formula (1):

[0068] (1)

[0069] Where: 、 : is the slip rate of the front and rear tires; 、 : Tire front and rear longitudinal stiffness; 、 : tire front and rear cornering stiffness; : longitudinal velocity of the vehicle, : lateral velocity of the vehicle, : yaw angle; : yaw angular velocity; : vehicle longitudinal acceleration; : vehicle lateral acceleration; : yaw angular acceleration; : The distance from the vehicle's center of mass to the front axle; : The distance from the vehicle's center of mass to the rear axle; : The moment of inertia of the vehicle around the z axis; m1: The mass of the vehicle; :The vehicle moves along the earth coordinate system Axis speed; The vehicle moves along the earth coordinate system The speed of the axis.

[0070] Then, based on the established three-degree-of-freedom model of the vehicle, the longitudinal velocity of the vehicle is selected , lateral speed , yaw angle , yaw angular velocity , and the vehicle's transverse and longitudinal positions are system state variables, as shown in formula (2):

[0071] (2)

[0072] The front wheel angle is the control quantity: First, the nonlinear dynamic model Perform linearization processing to obtain the linear state space equation, such as formula (3):

[0073] (3)

[0074] Where: 、 At the same time, the obtained linear state space equation is discretized to obtain formula (4):

[0075] (4)

[0076] Where: , m2 is the state dimension, T is the system sampling time. , we get the model expression of the prediction model, as shown in formula (5):

[0077] (5)

[0078] in, , , .

[0079] In this embodiment, the front wheel angle of the vehicle is used as the control output of the prediction model. At the same time, it is assumed that the vehicle speed remains unchanged. Considering that directly using the control output as the state quantity of the objective function may cause the control quantity to become larger, affecting the control accuracy, the increment of the control output is used as the state quantity of the objective function, as shown in formula (6):

[0080]

[0081] Where, is the optimal reference path; : prediction time domain; : Control time domain; : Tracking effect adjustment matrix; : Control quantity change adjustment matrix; : weight coefficient; : Relaxation factor to prevent the control quantity increment from having no solution.

[0082] During the solution process, the objective function must satisfy the following constraints:

[0083] Control output constraints:

[0084] (7)

[0085] Control output increment constraint:

[0086] (8)

[0087] Output variable constraints:

[0088] (9)

[0089] Considering that the road adhesion coefficient also constrains the vehicle's dynamics and directly affects the vehicle's acceleration, the specific relationship is:

[0090] In the formula 、 are the longitudinal acceleration and lateral acceleration of the vehicle, is the road adhesion coefficient. Assuming that the longitudinal speed of the vehicle remains constant, the above formula can be simplified to:

[0091]

[0092] Based on the above objective function and its constraints, it can be transformed into a quadratic programming problem through corresponding matrix operations, and the increment of the control output within the control domain can be solved. At the same time, the first increment is applied to the prediction model, and the above solution process is repeated until the solution process meets the constraints set by the objective function. The prediction model outputs the front wheel angle to achieve tracking of the optimal obstacle avoidance path.

[0093] A vehicle obstacle avoidance control method provided by an embodiment of the present invention fully considers the impact of the road adhesion coefficient on vehicle obstacle avoidance control, and thus constructs a driving risk field model for the road adhesion coefficient. Vehicle obstacle avoidance control based on this driving risk field model can greatly improve obstacle avoidance accuracy, thereby effectively reducing the driver's driving risk.

[0094] This embodiment provides another vehicle obstacle avoidance control method, which is implemented on the basis of the above embodiment; this embodiment focuses on the specific implementation methods of calculating the road adhesion coefficient, building a driving risk field model, and calculating the optimal reference path. Figure 2 The flowchart of another vehicle obstacle avoidance control method is shown in FIG. 1 . The vehicle obstacle avoidance control method in this embodiment includes the following steps:

[0095] S200, obtaining vehicle status information and road surface information during the current vehicle driving process;

[0096] S201, extracting vehicle tire information and vehicle speed information from vehicle status information in real time;

[0097] The above-mentioned vehicle tire information includes tire radius, tire angular velocity, tire lateral stiffness, tire longitudinal stiffness, tire lateral slip angle and other information, which can be collected by the sensors in the sensor group that collect the above-mentioned vehicle tire information.

[0098] The above-mentioned vehicle speed information includes vehicle speed information and acceleration information, which can be collected by the speed sensor and acceleration sensor in the sensor group.

[0099] S202, inputting vehicle tire information and vehicle speed information into a pre-trained tire model, and outputting tire adhesion coefficients corresponding to each tire of the vehicle through the tire model;

[0100] In this embodiment, the tire model is constructed as follows:

[0101] First, a three-degree-of-freedom vehicle dynamics model is constructed. The three-degree-of-freedom vehicle dynamics model is as shown in formula (10):

[0102]

[0103]

[0104] Where: is the front wheel turning angle; is the sideslip angle of the center of mass; is the yaw angular velocity; Distance from center of mass to front axle; Distance from center of mass to rear axle; lateral speed; longitudinal speed; Indicates the front; After indicating; Indicates left; Indicates right; 、 are longitudinal and lateral forces; For wheelbase.

[0105] The model formula for constructing the tire model based on the three-degree-of-freedom vehicle dynamics model is as follows (11):

[0106] (11)

[0107] in,

[0108] in,

[0109] in,

[0110] Where: is the tire radius; Tire angular velocity; Tire cornering stiffness; Tire longitudinal stiffness; Tire slip angle; Speed ​​impact factor; is the slip ratio; It is a nonlinear parameter used to describe tire slip; the subscript i takes f to represent the front axle and r to represent the rear axle; the subscript j takes l to represent the left wheel and r to represent the right wheel; for example, the subscript fl represents the left wheel on the front axle of the car, and rr represents the right wheel on the rear axle of the car.

[0111] In practice, methods for estimating the road adhesion coefficient are primarily categorized into two types: cause-based and effect-based. The cause-based algorithm requires specialized sensors, which is costly and susceptible to environmental influences, limiting its applicability. The effect-based algorithm estimates the road adhesion coefficient based on changes in vehicle motion parameters caused by road surface changes. This method offers low cost and wide applicability. Therefore, in this embodiment, the effect-based algorithm is used to calculate the road adhesion coefficient in real time, based on the functional relationship between the tire model and the road adhesion coefficient.

[0112] Specifically, the state equation and observation equation of the Effect-Based algorithm are established, and the state variables are selected as the tire adhesion coefficients of the four wheels:

[0113] In addition, assuming that the lateral and longitudinal accelerations 、 , yaw angular velocity It can be directly obtained from the sensor, so it is used as the observed variable, that is:

[0114] The control quantity is the front wheel angle and the four normalized tire forces:

[0115]

[0116] In summary, the state space equation based on the Effect-Based algorithm is as follows:

[0117] (12)

[0118] Where: : process noise; : Observation noise. Linearizing Equation (12), we can get the linearized state space expression as follows:

[0119]

[0120]

[0121] Where:

[0122]

[0123]

[0124]

[0125]

[0126] In summary, based on the state formula, observation formula, and control formula of the road adhesion coefficient estimation based on the Kalman filter and combined with the principle of the Kalman filter algorithm, the estimated tire adhesion coefficients of the vehicle's four tires can be obtained.

[0127] S203, calculating the average of multiple tire adhesion coefficients to obtain a road adhesion coefficient;

[0128] S204, extracting road boundary information, first position information of the vehicle heading towards the target position, current second position information of the obstacle, relative information between the obstacle and the vehicle, and current third position information of the vehicle from the road surface information;

[0129] Road boundary information can be extracted from road images captured by the camera in the sensor group. The determination of the position information of the above-mentioned first position information, second position information and third position information can be achieved by GPS positioning. The positioning accuracy of GPS is at the level of tens of meters to centimeters, and the positioning accuracy is high. Positioning can also be achieved by using a positioning method that integrates GPS and an inertial navigation system (Inertial Navigation System). The positioning method is not limited here.

[0130] The relative information between the obstacle and the vehicle includes relative speed information and acceleration information, which can be collected and obtained by the speed sensor and acceleration sensor.

[0131] S205, constructing a road boundary risk field model based on the road boundary information and the third position information;

[0132] Considering that most vehicles will drive close to the center line of the road when driving on the road, and this embodiment considers the vehicle lateral obstacle avoidance scenario, which is different from the vehicle lane change scenario, and ignores the influence of the lane line within the road boundary on the driving risk field model, therefore, a piecewise function is selected based on the road boundary information to model the road boundary risk field model, as shown in formula (13):

[0133] (13)

[0134] Where: The road boundary driving risk field adjustment coefficient is used to adjust the size of the road boundary risk field; and They are the left and right boundary positions of the road, namely the road boundary information.

[0135] Figure 3 A three-dimensional schematic diagram of a road boundary risk field model is shown. As can be seen from the figure, the driving risk is relatively small when the vehicle is driving normally within the road boundary. If the vehicle travels beyond the road boundary, the driving risk will increase. Therefore, an exponential function with a faster increasing speed is selected when the vehicle exceeds the road boundary, and the risk field value is zero within the road boundary.

[0136] S206, constructing a target gravitational field model based on the first position information and the third position information;

[0137] Considering that the role of the target gravity field model is to make the vehicle drive to the target position, the target gravity field model is as follows:

[0138] (14)

[0139] Where: is the target gravitational field adjustment coefficient; , x and y are the horizontal and vertical coordinates of the target position, i.e., the first position information; x and y are the horizontal and vertical coordinates of the vehicle's current position, i.e., the third position information.

[0140] Figure 4 A three-dimensional schematic diagram of a target gravitational field model is shown. Figure 4 It can be seen that the target gravity field model requires that the risk is greater at places far away from the target location and lower at places close to the target location, so that the gravity field tilts toward the target location and drives the vehicle to move toward the target location.

[0141] S207, constructing an obstacle risk field model based on the road adhesion coefficient, the second position information, the relative information, and the third position information;

[0142] Considering the requirement of smoothness of the vehicle's obstacle avoidance path, in this embodiment, a two-dimensional normal distribution function with a shape similar to that of the vehicle is selected to model the obstacle risk field. The model is as shown in formula (15):

[0143]

[0144] (15)

[0145] Where: : obstacle repulsion field size adjustment coefficient; : The relative speed between the test vehicle and the obstacle vehicle; : relative speed adjustment coefficient; : road adhesion coefficient; : Road adhesion coefficient adjustment coefficient; : Adjustment coefficient of the obstacle vehicle's overall dimensions; : relative acceleration adjustment coefficient; : relative acceleration.

[0146] in, , where 、 are the horizontal and vertical coordinates of the obstacle, i.e., the second position information.

[0147] Figure 5 A three-dimensional schematic diagram of obstacle risk field modeling is shown. Figure 5 It can be seen that the closer to the obstacle, the greater the risk value; and the purpose of adjusting the horizontal and vertical risk values ​​of the obstacle can be achieved by adjusting the major and minor axes of the obstacle vehicle risk field model.

[0148] S208, constructing a driving risk field model based on the road boundary risk field model, the target gravity field model, and the obstacle risk field model;

[0149] Specifically, a weighted calculation is performed on the road boundary risk field model, the target gravity field model, and the obstacle risk field model to obtain a driving risk field model.

[0150] In actual use, the weighting coefficient can be set according to actual needs and is not limited here. In this embodiment, the weighting coefficient is taken as 1 as an example, so the specific expression of the driving risk field model is: For ease of understanding, Figure 6 A three-dimensional schematic diagram of a driving risk field model is shown.

[0151] S209, performing negative gradient derivation on the driving risk field model to obtain multiple obstacle avoidance path points of the initial obstacle avoidance path;

[0152] The main purpose of vehicle obstacle avoidance is safety. Therefore, during the obstacle avoidance process, the vehicle should move in the direction where the risk value of the driving risk field decreases the fastest. According to relevant theories of advanced mathematics, when moving along the negative gradient direction, the function value decreases the fastest. By analogy with the driving risk field model, it can be seen that when the vehicle moves along the negative gradient direction of the driving risk field model, the risk value decreases the fastest and the obstacle avoidance safety is the highest. Therefore, the negative gradient derivative of the driving risk field model is performed to obtain multiple obstacle avoidance path points of the initial obstacle avoidance path.

[0153] S210, calculating an optimal reference path based on multiple obstacle avoidance path points;

[0154] Taking into account that the initial obstacle avoidance path planned directly by the negative gradient direction of the driving risk field model may be non-smooth and not conform to the vehicle dynamics constraints, multiple polynomial fittings are performed on multiple obstacle avoidance path points to obtain the optimal reference path. In this embodiment, a 5th-order polynomial fitting optimization can be used to obtain the optimal reference path.

[0155] S211, controlling the vehicle to track the optimal reference path to perform obstacle avoidance driving.

[0156] Corresponding to the above method embodiment, this embodiment provides a vehicle obstacle avoidance control device, wherein the device is applied to a vehicle-mounted server; see Figure 7 The schematic diagram of the structure of a vehicle obstacle avoidance control device shown in FIG. 1 includes:

[0157] An acquisition module 71 is used to acquire vehicle status information and road surface information during the current vehicle driving process;

[0158] A first calculation module 72 is used to calculate the road adhesion coefficient in real time based on the vehicle state information;

[0159] A construction module 73 is used to construct a driving risk field model based on the road adhesion coefficient and road surface information;

[0160] A second calculation module 74 is used to calculate the optimal reference path for obstacle avoidance based on the driving risk field model;

[0161] The control module 75 is used to control the vehicle to track the optimal reference path to avoid obstacles.

[0162] An embodiment of the present invention provides a vehicle obstacle avoidance control device. After obtaining vehicle status information and road surface information during the current vehicle driving process, the device calculates the road adhesion coefficient in real time based on the vehicle status information. A driving risk field model is constructed based on the road adhesion coefficient and road surface information. An optimal reference path for obstacle avoidance is calculated based on the driving risk field model, and the vehicle is controlled to track the optimal reference path for obstacle avoidance. This application fully considers the impact of the road adhesion coefficient on vehicle obstacle avoidance control and, therefore, constructs a driving risk field model for the road adhesion coefficient. Vehicle obstacle avoidance control based on this driving risk field model can significantly improve obstacle avoidance accuracy, thereby effectively reducing driving risks for the driver.

[0163] The first calculation module 72 is further configured to: extract vehicle tire information and vehicle speed information from the vehicle status information in real time; input the vehicle tire information and vehicle speed information into a pre-trained tire model, and output the tire adhesion coefficient corresponding to each tire of the vehicle through the tire model; and calculate the average of multiple tire adhesion coefficients to obtain a road adhesion coefficient.

[0164] The above-mentioned construction module 73 is also used to: extract road boundary information, first position information of the vehicle heading towards the target position, current second position information of the obstacle, relative information between the obstacle and the vehicle, and current third position information of the vehicle from the road surface information; construct a road boundary risk field model based on the road boundary information and the third position information; construct a target gravity field model based on the first position information and the third position information; construct an obstacle risk field model based on the road adhesion coefficient, the second position information, the relative information and the third position information; and construct a driving risk field model based on the road boundary risk field model, the target gravity field model and the obstacle risk field model.

[0165] The above-mentioned construction module 73 is further used to perform weighted calculation on the road boundary risk field model, the target gravity field model and the obstacle risk field model to obtain a driving risk field model.

[0166] The second calculation module 74 is further configured to perform negative gradient derivation on the driving risk field model to obtain a plurality of obstacle avoidance path points of the initial obstacle avoidance path; and calculate an optimal reference path based on the plurality of obstacle avoidance path points.

[0167] The second calculation module 74 is further configured to perform multiple multi-factor fitting calculations on multiple obstacle avoidance path points to obtain an optimal reference path.

[0168] An embodiment of the present invention provides a vehicle, which is equipped with an on-board controller, and the on-board controller is used to execute the above-mentioned vehicle obstacle avoidance control method.

[0169] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A vehicle obstacle avoidance control method, characterized in that: The method is applied to a vehicle-mounted controller; the method comprises: Obtain vehicle status information and road surface information during the current vehicle driving process; Calculating the road adhesion coefficient in real time based on the vehicle state information; Constructing a driving risk field model according to the road adhesion coefficient and the road information; Calculating an optimal reference path for obstacle avoidance based on the driving risk field model; Controlling the vehicle to track the optimal reference path to perform obstacle avoidance driving; The step of constructing a driving risk field model based on the road adhesion coefficient and the road surface information includes: Extracting road boundary information, first position information of the vehicle heading towards a target position, current second position information of an obstacle, relative motion information between the obstacle and the vehicle, and current third position information of the vehicle from the road surface information; constructing a road boundary risk field model based on the road boundary information and the third position information; constructing a target gravitational field model based on the first position information and the third position information; constructing an obstacle risk field model based on the road adhesion coefficient, the second position information, the relative motion information, and the third position information; A driving risk field model is constructed according to the road boundary risk field model, the target gravity field model and the obstacle risk field model.

2. The method according to claim 1, characterized in that The step of calculating the road adhesion coefficient in real time based on the vehicle state information includes: extracting vehicle tire information and vehicle speed information from the vehicle status information in real time; Inputting the vehicle tire information and the vehicle driving speed information into a pre-trained tire model, and outputting the tire adhesion coefficient corresponding to each tire of the vehicle through the tire model; The road adhesion coefficient is obtained by performing an average calculation on the plurality of tire adhesion coefficients.

3. The method according to claim 2, characterized in that The step of constructing a driving risk field model based on the road boundary risk field model, the target gravity field model, and the obstacle risk field model includes: A weighted calculation is performed on the road boundary risk field model, the target gravity field model, and the obstacle risk field model to obtain a driving risk field model.

4. The method according to claim 1, wherein The step of calculating the optimal reference path for obstacle avoidance based on the driving risk field model includes: Performing negative gradient derivation on the driving risk field model to obtain multiple obstacle avoidance path points of an initial obstacle avoidance path; An optimal reference path is calculated based on the plurality of obstacle avoidance path points.

5. The method according to claim 4, characterized in that The step of calculating the optimal reference path based on the plurality of obstacle avoidance path points comprises: Perform multiple multinomial fitting calculations on the plurality of obstacle avoidance path points to obtain an optimal reference path.

6. A vehicle obstacle avoidance control device, characterized in that: The device is applied to a vehicle-mounted server; The device comprises: An acquisition module is used to obtain vehicle status information and road surface information during the current vehicle driving process; A first calculation module, configured to calculate a road adhesion coefficient in real time based on the vehicle state information; A construction module, configured to construct a driving risk field model based on the road adhesion coefficient and the road surface information; A second calculation module is used to calculate an optimal reference path for obstacle avoidance based on the driving risk field model; A control module, configured to control the vehicle to track the optimal reference path to perform obstacle avoidance driving; Wherein, the building block is further used for: Extracting road boundary information, first position information of the vehicle heading towards a target position, second current position information of an obstacle, relative information between the obstacle and the vehicle, and third current position information of the vehicle from the road surface information; constructing a road boundary risk field model based on the road boundary information and the third position information; constructing a target gravitational field model based on the first position information and the third position information; constructing an obstacle risk field model based on the road adhesion coefficient, the second position information, the relative information, and the third position information; A driving risk field model is constructed according to the road boundary risk field model, the target gravity field model and the obstacle risk field model.

7. The device according to claim 6, characterized in that The first calculation module is further used to: extract vehicle tire information and vehicle speed information from the vehicle status information in real time; Inputting the vehicle tire information and the vehicle driving speed information into a pre-trained tire model, and outputting the tire adhesion coefficient corresponding to each tire of the vehicle through the tire model; The road adhesion coefficient is obtained by performing an average calculation on the plurality of tire adhesion coefficients.

8. A vehicle, characterized in that: The vehicle is equipped with an on-board controller, which is used to execute the vehicle obstacle avoidance control method described in any one of claims 1-5.

Images