Remote dynamic takeover method of vehicle, vehicle control assembly and processor
By predicting the driving trajectory and performing angle verification in unmanned vehicles, remote dynamic takeover without parking is achieved, the problem of inconsistent angle control parameters is solved, and the takeover efficiency and safety are improved.
Patent Information
- Application Number
- CN202411034888.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-30
- Publication Date
- 2024-11-08
AI Technical Summary
During the remote takeover of unmanned vehicles, due to the inconsistent angle control parameters between the autonomous driving system and the remote bridge, the steering wheel angle is inconsistent, and parking is required, which increases the cost of takeover time and poses safety risks, especially in emergency situations. It may cause traffic jams or increase the driving risk of other vehicles.
By querying the dynamic takeover request of the target vehicle, obtaining its status model, observation model and sensor data, predicting the driving trajectory and extracting the predicted value of the front wheel angle, performing angle verification, realizing remote dynamic takeover without parking, ensuring the steering wheel angle and The remote bridge angle is consistent.
It reduces the time cost during the takeover process, improves operating efficiency and safety, ensures vehicle stability and precise control, and avoids safety risks and traffic impacts during parking.
Smart Images

Figure CN118915551A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent transportation technology, and in particular to a remote dynamic takeover method for a vehicle, a vehicle control component and a processor. Background Art
[0002] During the driving process of an unmanned vehicle with automatic driving function and remote control and control function, when the remote driver takes over the vehicle, the angle control parameters between the automatic driving system and the remote driving console may be inconsistent, which may cause the steering wheel angle to be inconsistent. In order to ensure that the remote driver can take over the vehicle smoothly and keep consistent with the driving direction of the vehicle, it is necessary to achieve consistency by stopping the car.
[0003] However, the parking process will increase the time cost of the entire takeover process. Especially in emergency situations, you may need to take over the vehicle quickly to avoid potential dangers, such as avoiding obstacles or responding to sudden traffic conditions, which will also increase the safety risks of the parking process. In addition, on high-traffic roads or in congested conditions, the parking process may affect surrounding traffic, causing traffic jams or increasing the driving risks of other vehicles.
[0004] Therefore, how to provide a method for remote dynamic non-stop takeover control of an unmanned vehicle to overcome the shortcomings brought about by remote parking takeover of an unmanned vehicle is a technical problem that technical personnel in this field urgently need to solve. Summary of the invention
[0005] Based on the above problems, the present application provides a remote dynamic takeover method of a vehicle, a vehicle control component and a processor, which can overcome the shortcomings brought about by remote parking takeover of unmanned vehicles.
[0006] The embodiments of the present application disclose the following technical solutions:
[0007] A remote dynamic takeover method for a vehicle, applied to a vehicle control component, the method comprising:
[0008] Querying a dynamic takeover request of a target vehicle, wherein the driving mode of the target vehicle is an automatic driving mode;
[0009] In response to a dynamic takeover request of the target vehicle, acquiring a state model, an observation model, an initial state, and sensor data of the target vehicle;
[0010] Predicting a driving trajectory of the target vehicle based on a state model, sensor data, an observation model, and an initial state of the target vehicle;
[0011] Extracting a predicted value of a front wheel turning angle of the target vehicle from the driving trajectory, and performing a turning angle calibration based on the predicted value of the front wheel turning angle to obtain a calibration result;
[0012] Remote dynamic takeover of the vehicle is performed based on the verification result.
[0013] In a possible implementation, extracting a predicted value of a front wheel turning angle of the target vehicle from the driving trajectory, and performing a turning angle calibration based on the predicted value of the front wheel turning angle to obtain a calibration result, includes:
[0014] Acquire a predicted value of a front wheel turning angle of the target vehicle under the driving trajectory, and convert the predicted value of the front wheel turning angle of the target vehicle under the driving trajectory into a target steering wheel angle of the target vehicle;
[0015] Performing angle mapping on the target steering wheel angle to obtain a predicted mapping angle; the predicted mapping angle is a predicted value of a remote steering wheel angle of a remote driving console of the target vehicle;
[0016] Acquire an adjusted remote steering wheel angle of the remote driving station and an adjusted vehicle steering wheel angle of the target vehicle; the adjusted remote steering wheel angle of the remote driving station is obtained by adjusting according to the predicted mapping angle; the adjusted vehicle steering wheel angle of the target vehicle is obtained by adjusting according to the target steering wheel angle;
[0017] Subtracting the adjusted remote steering wheel angle of the remote driving station from the adjusted vehicle steering wheel angle of the target vehicle to obtain an angle error;
[0018] A rotation angle check is performed based on the angle error and the rotation angle rule to obtain a check result.
[0019] In a possible implementation, the remote dynamic takeover of the vehicle based on the verification result includes:
[0020] When the verification result is successful, controlling the target vehicle to exit the automatic driving mode and remotely dynamically taking over the vehicle;
[0021] When the verification result is in verification, the step of predicting the driving trajectory of the target vehicle and subsequent steps based on the state model, sensor data, observation model and initial state of the target vehicle is performed again;
[0022] When the verification result is a verification failure, the target vehicle is controlled to maintain the automatic driving mode, and the dynamic takeover request of the target vehicle is queried again.
[0023] In a possible implementation, the corner rule includes:
[0024] When the angle error is less than the preset error, the verification result is fed back as verification success;
[0025] When the angle error is greater than or equal to the preset error, the verification result is fed back as being in verification;
[0026] When the verification result is fed back as being in verification, the steps of obtaining the state model, observation model, initial state and sensor data of the target vehicle and subsequent steps are executed again to obtain a new angle error. When the new angle error is still greater than or equal to the preset error within a preset time period, the verification result is changed to verification failure; the preset time period refers to the time period from obtaining the state model, observation model, initial state and sensor data of the target vehicle to a preset time; and / or,
[0027] When the verification result is fed back as verification in progress, the steps of obtaining the state model, observation model, initial state and sensor data of the target vehicle and subsequent steps are executed again to obtain a new angle error. When the new angle error is still greater than or equal to the preset error within a preset number of times, the verification result is changed to verification failure. The preset number of times is the number of times the steps of obtaining the state model, observation model, initial state and sensor data of the target vehicle and subsequent steps are executed again to obtain a new angle error.
[0028] In a possible implementation, the corner rule further includes:
[0029] During the angle verification process, the adjusted remote steering wheel angle of the remote driving station changes, and the verification result is fed back as verification failure; and / or,
[0030] During the turning angle verification process, the remote brake pedal of the target vehicle is subjected to force, and the verification result is fed back as verification failure; and / or,
[0031] During the turning angle verification process, a remote driving pedal of the target vehicle is subjected to force, and the verification result is fed back as verification failure.
[0032] In a possible implementation, predicting the driving trajectory of the target vehicle based on the state model, observation model, initial state and sensor data of the target vehicle includes:
[0033] Based on the state model, observation model, initial state and sensor data of the target vehicle, predict the predicted position of the target vehicle at the next moment by using an extended Kalman filter (EKF) algorithm;
[0034] The driving trajectory of the target vehicle is predicted using a model predictive control (MPC) model based on the predicted position and trajectory constraints.
[0035] In a possible implementation, the trajectory constraint condition includes:
[0036] The front wheel steering angle is greater than or equal to -10 degrees and less than or equal to 10 degrees; and / or,
[0037] The front wheel steering angle changes by greater than or equal to -0.8 degrees and less than or equal to 0.8 degrees; and / or
[0038] The front wheel slip angle is greater than or equal to -2.8 degrees and less than or equal to 2.8 degrees; and / or,
[0039] The rear wheel slip angle is greater than or equal to -2.8 degrees and less than or equal to 2.8 degrees.
[0040] In a possible implementation, the method further includes:
[0041] When the verification result is successful, the user is prompted that the target vehicle exits the automatic driving mode.
[0042] A vehicle control component, the vehicle control component comprising a remote driving console, a remote driving module and an automatic driving module;
[0043] The remote driving console is used to receive a dynamic takeover request of a target vehicle sent by a remote driver; wherein the driving mode of the target vehicle is an automatic driving mode.
[0044] The remote driving module is used to process the dynamic takeover request of the target vehicle and forward the processed dynamic takeover request of the target vehicle to the automatic driving module;
[0045] The autonomous driving module is used to query the dynamic takeover request of the target vehicle after information processing forwarded by the remote driving module;
[0046] In response to querying the dynamic takeover request of the target vehicle after information processing forwarded by the remote driving module, the automatic driving module is further used to obtain the state model, observation model, initial state and sensor data of the target vehicle, and predict the driving trajectory of the target vehicle based on the state model, observation model, initial state and sensor data of the target vehicle;
[0047] The automatic driving module further extracts a predicted value of a front wheel turning angle of the target vehicle from the driving trajectory;
[0048] The remote driving module is further used to perform a turning angle calibration based on the predicted value of the front wheel turning angle to obtain a calibration result, and to perform remote dynamic takeover of the vehicle based on the calibration result.
[0049] A processor is used to run a computer program, which executes the remote dynamic takeover method of a vehicle as described above when running.
[0050] Compared with the prior art, this application has the following beneficial effects:
[0051] The present application provides a remote dynamic takeover method for a vehicle, a vehicle control component and a processor. Specifically, when executing the remote dynamic takeover method for a vehicle provided in an embodiment of the present application, the dynamic takeover request of the target vehicle whose driving mode is the automatic driving mode can be queried first, and the state model, observation model, initial state and sensor data of the target vehicle can be obtained when the dynamic takeover request of the target vehicle is queried. Then, based on the state model, observation model, initial state and sensor data of the target vehicle, the driving trajectory of the target vehicle is predicted, and the predicted value of the front wheel turning angle of the target vehicle contained in the driving trajectory is extracted. Then, the turning angle is verified based on the predicted value of the front wheel turning angle to obtain the verification result, and the remote dynamic takeover of the vehicle is performed based on the verification result. Compared with the method of unifying the angle control parameters by stopping, the present application can realize the remote dynamic takeover of the vehicle without stopping. This can reduce the time cost in the takeover process and improve the overall operation efficiency. At the same time, by predicting the driving trajectory of the target vehicle and extracting the front wheel turning angle, the state change of the target vehicle can be responded to more in real time, and timely adjustments can be made during the takeover process to ensure the safety and stability of the vehicle. In addition, by predicting the driving trajectory and extracting the front wheel turning angle, the vehicle's driving direction and posture can be controlled more accurately, improving the accuracy and reliability of remote takeover. BRIEF DESCRIPTION OF THE DRAWINGS
[0052] In order to more clearly illustrate the technical solutions in this embodiment or 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0053] Figure 1 A schematic diagram of an exemplary application scenario provided for an embodiment of the present application;
[0054] Figure 2 A method flow chart of a method for remote dynamic takeover of a vehicle provided in an embodiment of the present application;
[0055] Figure 3 A flow chart of a method for performing a turning angle calibration based on a predicted value of a front wheel turning angle provided in an embodiment of the present application;
[0056] Figure 4 A schematic diagram of the structure of a vehicle control component provided in an embodiment of the present application. DETAILED DESCRIPTION
[0057] To facilitate understanding of the technical solutions provided by the embodiments of the present application, the background technology involved in the embodiments of the present application will be described below.
[0058] When unmanned vehicles have autonomous driving and remote takeover functions, the remote driver may face challenges caused by inconsistent angle control parameters when taking over the vehicle. In order to ensure that the remote driver takes over the vehicle smoothly and keeps consistent with the vehicle's driving direction, it is usually necessary to unify the angle control parameters by parking the vehicle. However, parking will increase the time cost of the takeover process, especially in emergency situations where it may be necessary to take over the vehicle quickly to avoid potential dangers, such as avoiding obstacles or dealing with sudden traffic conditions, which will also increase the safety risks of the parking process.
[0059] In addition, on high-traffic roads or in congested conditions, the parking process may have an impact on surrounding traffic, possibly causing traffic jams or increasing the driving risks of other vehicles.
[0060] In order to solve this problem, an embodiment of the present application provides a method for remote dynamic takeover of a vehicle, a vehicle control component and a processor. First, the dynamic takeover request of the target vehicle is queried, and the state model, observation model, initial state and sensor data of the target vehicle are obtained when the dynamic takeover request of the target vehicle is queried. Then, based on the state model, observation model, initial state and sensor data of the target vehicle, the driving trajectory of the target vehicle is predicted, and the predicted value of the front wheel angle of the target vehicle is extracted from the driving trajectory. Next, the angle is verified based on the predicted value of the front wheel angle to obtain a verification result, so as to remotely and dynamically take over the vehicle based on the verification result. Compared with the method of unifying the angle control parameters that requires parking, the present application can realize remote dynamic takeover of the vehicle without parking. This can reduce the time cost in the takeover process and improve the overall operating efficiency.
[0061] In order to facilitate understanding of the remote dynamic takeover method for a vehicle provided in the embodiment of the present application, Figure 1 See the example scenario shown in the figure. Figure 1 , which is a framework diagram of an exemplary application scenario provided in an embodiment of the present application.
[0062] When the target vehicle in autonomous driving mode encounters complex road conditions during driving, such as obstacles in front of the vehicle's path, vehicles driving in lanes that can be changed, etc., in order to protect the safety and continuity of the unmanned vehicle's driving, the remote driver needs to take over dynamically. At this time, the remote driver will control the remote driving console and send a dynamic takeover request to the target vehicle through the remote driving module. The autonomous driving module cannot recognize the dynamic takeover request received by the remote driving console, so it is necessary to process the dynamic takeover request received by the remote driving console through the remote driving module and convert the dynamic takeover request into information that can be recognized by the autonomous driving module. A remote driver, also known as a remote operator or remote driving expert, is a person who monitors and controls an unmanned vehicle through remote communication technology.
[0063] When the automatic driving module in the vehicle control component of the target vehicle receives the dynamic takeover request of the target vehicle forwarded by the remote driver through the remote driving module, it will obtain the state model, observation model, initial state and sensor data of the target vehicle, so as to predict the driving trajectory of the target vehicle at the next moment based on the state model, observation model, initial state and sensor data of the target vehicle. The state model is used to describe the motion law of the target vehicle within a fixed time interval, including the evolution equations of state variables such as position, speed, acceleration, etc. Sensor data refers to the actual collected sensor data, such as GPS data, lidar data, camera images, etc. The observation model is used to describe the relationship between the sensor data of the target vehicle and the state model. The initial state refers to the estimation of the initial state of the vehicle, including information such as position, speed, direction, etc. The automatic driving module refers to a hardware and software system installed on the vehicle, which has autonomous perception, decision-making and control capabilities, and can realize the function of autonomous driving of the vehicle.
[0064] The autonomous driving module then extracts the predicted value of the target vehicle's front wheel angle from the driving trajectory, converts the predicted value into the target steering wheel angle of the target vehicle, and then sends the target steering wheel angle to the remote driving module.
[0065] After receiving the target steering wheel angle, the remote driving module maps the target steering wheel angle, obtains the predicted value of the remote steering wheel angle of the remote driving console of the target vehicle, i.e., the predicted mapping angle, and sends the predicted mapping angle to the remote driving console, which adjusts the remote steering wheel angle through a control device. A remote driving console refers to a device that monitors and controls unmanned vehicles through remote communication technology. A remote driving console usually consists of a computer, a display, a camera, etc. The remote driving module is usually used in conjunction with the automatic driving module and the remote driving console to realize remote monitoring and control of vehicle driving.
[0066] The remote driving module then obtains the actual remote steering wheel angle adjusted according to the predicted mapping angle, and simultaneously obtains the actual vehicle steering wheel angle adjusted according to the target steering wheel angle. During the process of adjusting the remote steering wheel angle according to the predicted mapping angle, the steering wheel of the target vehicle will change its angle accordingly, thereby obtaining a new vehicle steering wheel angle.
[0067] Finally, the remote driving module performs remote dynamic takeover of the vehicle based on the actual remote steering wheel angle and the actual vehicle steering wheel angle. In the process of remote dynamic takeover of the vehicle based on the actual remote steering wheel angle and the actual vehicle steering wheel angle, it is necessary to subtract the actual remote steering wheel angle from the actual vehicle steering wheel angle to obtain the angle error. Angle verification is performed based on the angle error and the angle rule to obtain the verification result, and the remote dynamic takeover of the vehicle is performed according to the verification result. When the verification result is successful, the automatic driving module will exit the vehicle control (that is, the vehicle exits the automatic driving mode), and the vehicle responds to the remote driving module vehicle control command to achieve remote vehicle control. In other words, during the entire remote dynamic takeover of the vehicle, there will always be information transmission, that is, the automatic driving module will always send vehicle information to the vehicle, and the remote driving module will always send some vehicle control information from the driving module to the vehicle, but when the vehicle's driving mode is the automatic driving mode, the vehicle does not respond to the remote driving command. Only when the vehicle exits the automatic driving mode will the vehicle respond to the remote driving module's vehicle control command.
[0068] It should be noted that in the process of the remote driving module taking over the vehicle remotely and dynamically based on the actual remote steering wheel angle and the actual vehicle steering wheel angle, it is necessary to "perform angle verification based on angle error and angle rules". The angle verification can be completed by the remote driving module or by the automatic driving module.
[0069] This enables dynamic remote takeover of the vehicle without stopping, thereby reducing the time cost during the takeover process and improving overall operational efficiency.
[0070] Those skilled in the art will understand that Figure 1 The framework diagram shown is only an example in which the embodiments of the present application can be implemented. The scope of application of the embodiments of the present application is not limited in any aspect by the framework.
[0071] 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 application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0072] See also Figure 2 , which is a flow chart of a method for remote dynamic takeover of a vehicle provided in an embodiment of the present application, such as Figure 2 As shown, the remote dynamic takeover method of the vehicle may include steps S201-S205:
[0073] S201: Query a dynamic takeover request of a target vehicle, wherein the driving mode of the target vehicle is an automatic driving mode.
[0074] When the target vehicle in autonomous driving mode encounters complex road conditions during driving, such as obstacles in front of the vehicle and vehicles driving in the lane where the vehicle can change lanes, the remote driver needs to take over dynamically to protect the safety and continuity of the unmanned vehicle. At this time, the remote driver will send a dynamic takeover request to the control component of the target vehicle.
[0075] It should be noted that during the dynamic takeover of vehicle remote driving, the driver needs to check the driving mode of the target vehicle, and only when the driving mode of the target vehicle is the automatic driving mode, will a dynamic takeover request be issued to the control component of the target vehicle.
[0076] The vehicle control components include a remote driving module, an automatic driving module, and a remote driving console. A remote driving console is a device that monitors and controls unmanned vehicles through remote communication technology. A remote driving console usually consists of a computer, a display, a camera, etc. A remote driving module is a software and hardware system that can remotely control a vehicle. An automatic driving module is a hardware and software system installed on a vehicle that has autonomous perception, decision-making, and control capabilities and can realize the function of autonomous driving of the vehicle. A remote driving module is usually used in conjunction with an automatic driving module and a remote driving console to remotely monitor and control vehicle driving.
[0077] A remote driver, also known as a remote operator or remote driving expert, is a person who monitors and controls an unmanned vehicle through remote communication technology.
[0078] S202: In response to the dynamic takeover request of the target vehicle being queried, a state model, an observation model, an initial state and sensor data of the target vehicle are acquired.
[0079] When the vehicle control component of the target vehicle receives a dynamic takeover request of the target vehicle sent by the remote driver, it will promptly obtain the state model, observation model, initial state and sensor data of the target vehicle.
[0080] The state model is used to describe the motion law of the target vehicle within a fixed time interval, including the evolution equations of state variables such as position, velocity, acceleration, etc. This is usually a nonlinear state equation that describes the state changes of the vehicle during motion.
[0081] Sensor data refers to the actual collected sensor data, such as GPS data, lidar data, camera images, etc. This data is used to update state estimation and correct prediction results.
[0082] The observation model is used to describe the relationship between the sensor data of the target vehicle and the state model. The observation model can also be nonlinear, and the sensor data needs to be converted into state space in order to be fused with the state model.
[0083] The initial state refers to the estimate of the initial state of the vehicle, including information such as position, speed, direction, etc. This initial state is used to initialize the algorithm and provide a starting point when there is no actual observation data.
[0084] S203: Predicting the driving trajectory of the target vehicle based on the state model, sensor data, observation model and initial state of the target vehicle.
[0085] By leveraging the state model, sensor data, observation model, and initial state, the vehicle control component can predict the future trajectory of the target vehicle and prepare for takeover.
[0086] In a possible implementation, predicting the driving trajectory of the target vehicle based on the state model, observation model, initial state and sensor data of the target vehicle includes:
[0087] Based on the state model, observation model, initial state and sensor data of the target vehicle, the predicted position of the target vehicle at the next moment is predicted by the EKF (Extend Kalman Filter) algorithm. Then, based on the predicted position and trajectory constraints, the MPC (Model Predictive Control) model is used to predict the driving trajectory of the target vehicle.
[0088] Specifically, since the processing of positioning error and noise also has a certain impact on positioning accuracy, the EKF algorithm is first used to predict the vehicle's driving trajectory. The formula for calculating the vehicle state using the EKF algorithm is shown in the following formula (1):
[0089]
[0090] Among them, x t is the state vector of the vehicle at time t; u is the control vector, such as acceleration, angular acceleration, etc.; Z tis the state vector of the sensor measurement value at time t, that is, the measurement result of the sensor; W is the process evolution noise; V is the measurement noise, f() is the state function, and h() is the measurement function.
[0091] The state of the vehicle at time t is Among them, (x, y) is the vehicle position, v is the vehicle speed, ω is the vehicle angular velocity, and θ is the vehicle heading angle.
[0092] After time △t, the state of the vehicle is
[0093] When ω≠0,
[0094]
[0095] When ω=0,
[0096]
[0097] Specifically, when predicting the driving trajectory of a vehicle, the roadside server can predict the position of the vehicle at the next moment according to the prediction model shown in the following formula (2).
[0098]
[0099] Among them, x t+1 is the position of the vehicle at time t+1, x t is the position of the vehicle at time t, g() function is the vehicle state transfer function, u is the control vector, such as acceleration; P t+1 is the covariance matrix corresponding to the vehicle state at time t+1, P t is the covariance matrix corresponding to the vehicle state at time t, J A is the Jacobian matrix, and Q is the covariance matrix of the Gaussian noise of the predicted state.
[0100] When ω≠0, J A As shown in the following formula (3):
[0101]
[0102] When ω=0, J A As shown in the following formula (4):
[0103]
[0104] Among them, ω is the vehicle angular velocity, θ is the vehicle heading angle, v is the vehicle speed, and △t is the time difference between two adjacent moments.
[0105] The covariance matrix Q is shown in the following formula (5):
[0106]
[0107] Among them, σ a is the acceleration noise on the straight line, σ ω is the angular velocity noise on the corner.
[0108] In some possible implementations, after the vehicle's position is predicted, the uncertainty of the vehicle's position is increasing. Therefore, the vehicle needs to measure the environment and acquire data in order to update its state and reduce the uncertainty of its position.
[0109] In some possible implementations of the embodiments of the present application, the vehicle position is updated according to the following formula (6).
[0110] x′ t =x t +K t (z t -h(x t )) (6)
[0111] Among them, x′ t is the updated vehicle position, x t is the vehicle position before updating, K t is the Kalman gain corresponding to time t, Z t is the state vector of the sensor measurement value at time t, and h() is the measurement function.
[0112] K t The formula is shown in formula (7):
[0113]
[0114] Where R is the covariance matrix of the Gaussian noise of the measurement value, that is, the measurement error of the sensor.
[0115] J H The formula is shown in formula (8):
[0116]
[0117] Among them, (x, y) is the vehicle position and v is the vehicle speed.
[0118] P t The formula is shown in formula (9):
[0119] P t =(IK t J H ) t (9)
[0120] Among them, P t is the state estimation covariance matrix, and I is the identity matrix.
[0121] The vehicle state corrected by the EKF algorithm is used for optimization of the MPC model predictive control.
[0122] Specifically, a vehicle dynamics model must first be established to describe the vehicle trajectory. To facilitate calculation, an idealized two-degree-of-freedom vehicle dynamics model is established as shown in the following formula (10):
[0123]
[0124] Where (x, y) is the vehicle position; is the vehicle's lateral and longitudinal speed; is the vehicle's lateral and longitudinal acceleration; m is the vehicle's mass; Φ is the yaw angle; is the yaw angular velocity; δ f is the front wheel turning angle of the car; I z F is the moment of inertia of the vehicle around the vertical axis z-axis; lf Front tire longitudinal force; F lr is the rear tire longitudinal force; F cf Front tire lateral force; F cr is the rear tire lateral force; a is the distance from the center of mass of the vehicle to the front axle; b is the distance from the center of mass of the vehicle to the rear axle.
[0125] The model is simplified, and the longitudinal force Flf, Flr and the lateral force Fcf, Fcr of the tire in the vehicle dynamics equation are optimized by combining the linearized tire function, as shown in the following formula (11):
[0126]
[0127] Among them, C lf Front tire longitudinal stiffness; C lr is the longitudinal stiffness of the rear tire; C cf Front tire cornering stiffness; C cr is the rear tire cornering stiffness; s lf Front tire slip rate; s lr is the rear tire slip rate; a cf Front tire slip angle; a cr is the rear tire slip angle.
[0128] Combining formulas (10) and (12), the simplified vehicle dynamics nonlinear model is shown in formula (12):
[0129]
[0130] The state quantity is The control amount is μ = δ f .
[0131] Next, based on the current state and control variables, the MPC model is used to predict the vehicle system output in the future time domain. First, the linear time-varying state equation is introduced as shown in formula (13):
[0132]
[0133] Where A(t) and B(t) are the Jacobian matrices of the state equation corresponding to ξ and μ; the system output is η(t) = [Y, φ]; the output matrix is C = [001010] T .
[0134] Secondly, formula (13) is discretized and the incremental model is introduced. The state space expression after processing is shown in formula (14):
[0135]
[0136] Among them, A(k)=I+TA(t); B(k)=TB(t); I is the unit matrix; T is the sampling period.
[0137] Next, the MPC model is used for prediction.
[0138]
[0139] in,
[0140] The system input and output at time t in the time domain are shown in formula (16):
[0141] Y(t+1|t)=[y(t+1|t), y(t+2|t),…,y(t+N P |t)] T
[0142] ΔU(t)=[Δu(t|t), Δu(t+1|t),·…, Δu(t+NC|t)] T (16)
[0143] Among them, N p is the prediction time domain; N C To control the time domain.
[0144] In order to avoid dangerous situations such as vehicle rollover due to excessive turning angles during actual driving, the MPC controller needs to be constrained in trajectory. The trajectory constraint conditions can be selected accordingly according to the actual situation.
[0145] In a possible implementation, the trajectory constraint condition includes:
[0146] Front wheel turning angle δ fgreater than or equal to -10 degrees and less than or equal to 10 degrees; and / or,
[0147] Front wheel turning angle change Δδ f Greater than or equal to -0.8 degrees and less than or equal to 0.8 degrees; and / or
[0148] Front wheel slip angle a cf greater than or equal to -2.8° degrees and less than or equal to 2.8 degrees; and / or,
[0149] Rear wheel slip angle a cr Greater than or equal to -2.8° degrees and less than or equal to 2.8 degrees.
[0150] It should be noted that the difference between the front wheel angle and the steering wheel angle is the "angle transmission ratio". For example, the steering wheel can generally only be turned to a maximum of 540 degrees, and the maximum wheel angle is basically only about 30 degrees.
[0151] Next, the control sequence in the future time domain is solved through the objective function, and the relaxation factor is introduced to avoid the situation where there is no solution. The objective function is shown in formula (17):
[0152]
[0153] Among them, ρ is the weight coefficient; ε is the relaxation factor; Q and R are weight matrices; Δy((t+i)|t) is the difference between the reference trajectory and the system output; Δu((t+i)|t) is the system control increment; i is the order size.
[0154] The MPC controller solves the optimal value based on the objective function and the introduced constraints: minJ(ξ(t), u(t-1), △u(t)).
[0155] After solving in each prediction time domain Np, the corresponding MPC system control input increment Δut* and relaxation factor ε in the control time domain NC are obtained: △u t * =[△u t * , △u t+1 * , ..., △u t+Nc-1 * , ε].
[0156] Combined with formula (17), using the first element in the formula as the actual control input increment, calculate the predicted output: u(t) = u(t-1) + △u t * , and then repeat the above steps to solve the problem cyclically to realize the trajectory tracking control of MPC algorithm.
[0157] S204: extracting a predicted value of a front wheel turning angle of the target vehicle from the driving trajectory, and performing a turning angle verification based on the predicted value of the front wheel turning angle to obtain a verification result.
[0158] Based on the predicted driving trajectory, the vehicle control component extracts the predicted value of the front wheel steering angle of the target vehicle in this trajectory to ensure that the vehicle follows the expected path. The driving trajectory data may include information such as vehicle position, direction, and speed. The vehicle control component will perform a steering check on the predicted front wheel steering angle value. After the verification process is completed, a verification result will be obtained, which can indicate whether the predicted front wheel steering angle meets the system's expected standards.
[0159] Alternatively, based on the determined driving trajectory, the path tracking control algorithm can be used to extract the turning angle that the front wheels of the vehicle should take under the trajectory. This turning angle will make the vehicle drive along the planned trajectory while meeting the vehicle's stability and handling performance requirements.
[0160] Optionally, the verification result may be a Boolean value (compliant / noncompliant) or a quantitative evaluation value.
[0161] See also Figure 3 , Figure 3 A flow chart of a method for performing a turning angle calibration based on a predicted value of a front wheel turning angle provided in an embodiment of the present application. Accordingly, in step S204, a predicted value of the front wheel turning angle of the target vehicle is extracted from the driving trajectory, and a turning angle calibration is performed based on the predicted value of the front wheel turning angle to obtain a calibration result, which can be specifically implemented through steps A1-A5:
[0162] A1: Obtain a predicted value of a front wheel turning angle of the target vehicle under the driving trajectory, and convert the predicted value of the front wheel turning angle of the target vehicle under the driving trajectory into a target steering wheel angle of the target vehicle.
[0163] By analyzing the target vehicle's motion state, road conditions, environmental factors, etc. under the predicted driving trajectory, the target vehicle's front wheel should be predicted. This predicted value can help the remote driver understand how the vehicle should turn to drive along a specific trajectory.
[0164] The predicted front wheel angle value is converted into a target steering wheel angle, that is, the angle to which the remote driver needs to turn the steering wheel to achieve the desired steering effect. This conversion process may involve the connection between the vehicle's steering mechanism, control system, and remote driving console to ensure that the remote driver can accurately control the steering of the target vehicle.
[0165] A2: Performing angle mapping on the target steering wheel angle to obtain a predicted mapping angle.
[0166] The predicted mapping angle is a predicted value of a remote steering wheel angle of a remote driving console of the target vehicle.
[0167] The predicted value of the steering wheel angle of the target vehicle under the driving trajectory is mapped to a steering wheel angle of the actual vehicle (i.e., the target vehicle) under the driving trajectory and a steering wheel angle of the remote driving console. Mapping the predicted value of the actual steering wheel angle of the target vehicle to the steering wheel angle of the remote driving console can facilitate the remote driver to accurately control the vehicle.
[0168] For example, assuming that the steering wheel angle of the actual vehicle ranges from -30 degrees to 30 degrees, and the steering wheel angle of the remote console ranges from -450 degrees to 450 degrees. If the predicted steering wheel angle of the target vehicle under the driving trajectory is 25 degrees, a linear mapping can be used to map the predicted steering wheel angle to the angle range of the remote console.
[0169] First, the proportional relationship between the two ranges needs to be determined. Within the range of the actual vehicle, the range of the steering wheel angle is 60 degrees (30 degrees - (-30 degrees) = 60 degrees). Within the range of the remote driving station, the range of the steering wheel angle is 900 degrees (450 degrees - (-450 degrees) = 900 degrees).
[0170] Then, the following linear mapping formula can be used for corner mapping:
[0171] Mapping angle = (target angle - actual angle lower limit) * (remote control station angle range / actual angle range) + remote control station angle lower limit
[0172] Substitute the specific values into the formula to calculate the angle of the mapping angle:
[0173] Mapping angle = (25 degrees - (-30 degrees)) * (900 degrees / 60 degrees) + (-450 degrees)
[0174] =(55 degrees)*(15)+(-450 degrees)
[0175] =825 degrees + (-450 degrees)
[0176] =375 degrees
[0177] Therefore, according to the mapping conditions, when the predicted steering wheel angle of the target vehicle under the driving trajectory is 25 degrees, the corresponding predicted mapping angle is 375 degrees.
[0178] A3: Obtaining the adjusted remote steering wheel angle of the remote driving console and the adjusted vehicle steering wheel angle of the target vehicle.
[0179] The adjusted steering wheel angle of the remote control console is obtained by adjusting the predicted mapping angle. The adjusted vehicle steering wheel angle of the target vehicle is obtained by adjusting the target steering wheel angle.
[0180] A4: Subtract the adjusted remote steering wheel angle of the remote driving station from the adjusted vehicle steering wheel angle of the target vehicle to obtain an angle error.
[0181] The difference between the adjusted remote steering wheel angle of the remote driving console and the adjusted vehicle steering wheel angle of the target vehicle is calculated as the angle error.
[0182] A5: Perform an angle check based on the angle error and the angle rule to obtain a check result.
[0183] Corner verification refers to verification based on the angular error between the vehicle's current state and target state, combined with pre-set corner rules, to ensure that the vehicle can turn according to the expected trajectory and control requirements during driving.
[0184] In a possible implementation, the corner rule includes:
[0185] (1) When the angle error is less than a preset error, the verification result is fed back as verification success.
[0186] (2) When the angle error is greater than or equal to a preset error, the verification result is fed back as being in verification.
[0187] (3) when the feedback of the verification result is that the verification is in progress, the steps of obtaining the state model, observation model, initial state and sensor data of the target vehicle and subsequent steps are executed again to obtain a new angle error, and when the new angle error is still greater than or equal to the preset error within a preset time period, the verification result is changed to verification failure; the preset time period refers to the time period from the acquisition of the state model, observation model, initial state and sensor data of the target vehicle to the preset time; and / or,
[0188] (4) When the feedback of the verification result is that the verification is in progress, the steps of obtaining the state model, observation model, initial state and sensor data of the target vehicle and subsequent steps are executed again to obtain a new angle error. When the new angle error is still greater than or equal to the preset error within a preset number of times, the verification result is changed to verification failure. The preset number of times is the number of times the steps of obtaining the state model, observation model, initial state and sensor data of the target vehicle and subsequent steps are executed again to obtain a new angle error.
[0189] In a possible implementation, the preset time and the preset error are of the same value, and the present application does not impose any specific limitation on the values of the preset error and the preset time.
[0190] In a possible implementation, the corner rule further includes:
[0191] During the angle verification process, the adjusted remote steering wheel angle of the remote driving station changes, and the verification result is fed back as verification failure; and / or,
[0192] During the turning angle verification process, the remote brake pedal of the target vehicle is subjected to force, and the verification result is fed back as verification failure; and / or,
[0193] During the turning angle verification process, a remote driving pedal of the target vehicle is subjected to force, and the verification result is fed back as verification failure.
[0194] That is to say, the remote driver is prohibited from operating the steering wheel, stepping on the brake pedal or the drive pedal during the calibration process, otherwise it will be considered that the calibration has failed.
[0195] S205: Performing remote dynamic takeover of the vehicle based on the verification result.
[0196] The vehicle control component will realize remote dynamic takeover of the target vehicle based on the verification results, that is, control the driving direction and path of the target vehicle through the remote operator to ensure the safe driving of the vehicle.
[0197] In a possible implementation, the remote dynamic takeover of the vehicle based on the verification result includes:
[0198] When the verification result is successful, the target vehicle is controlled to exit the automatic driving mode and remote dynamic takeover of the vehicle is performed.
[0199] If the verification is successful, the target vehicle's autonomous driving mode will be exited. At the same time, the remote driver is allowed to remotely take over the controlled vehicle, and the unmanned vehicle is in remote control mode.
[0200] When the verification result is in process, the step of predicting the driving trajectory of the target vehicle and subsequent steps based on the state model, sensor data, observation model and initial state of the target vehicle is executed again.
[0201] If the verification result is "verifying", the target vehicle's driving trajectory needs to be predicted again and the predicted value is sent to the remote driving console. The remote driving console automatically adjusts and sends a new angle value again, and verifies again based on the new input.
[0202] When the verification result is a verification failure, the target vehicle is controlled to maintain the automatic driving mode, and the dynamic takeover request of the target vehicle is queried again.
[0203] If the verification fails, the target vehicle will continue to maintain the autonomous driving mode and send the verification results to the remote driving console so that the remote driver can determine whether to send a dynamic takeover request again based on the actual situation.
[0204] In a possible implementation, the method further includes:
[0205] When the verification result is successful, the user is prompted that the target vehicle exits the automatic driving mode.
[0206] Based on the contents of S201-S205, it can be known that the dynamic takeover request of the target vehicle is queried first, wherein the driving mode of the target vehicle is the automatic driving mode. When the dynamic takeover request of the target vehicle is queried, the state model, observation model, initial state and sensor data of the target vehicle are obtained. Then, based on the state model, observation model, initial state and sensor data of the target vehicle, the driving trajectory of the target vehicle is predicted, and the predicted value of the front wheel turning angle of the target vehicle is extracted from the driving trajectory, and the angle verification is performed based on the predicted value of the front wheel turning angle to obtain the verification result. Finally, the angle verification is performed based on the predicted value of the front wheel turning angle to obtain the verification result, and the remote dynamic takeover of the vehicle is performed based on the verification result. Compared with the method of unifying the angle control parameters that requires parking, the present application can realize the remote dynamic takeover of the vehicle without parking, so as to reduce the time cost in the takeover process and improve the overall operation efficiency. At the same time, by predicting the driving trajectory of the target vehicle and extracting the front wheel turning angle, the state change of the target vehicle can be responded to more in real time, and timely adjustments can be made during the takeover process to ensure the safety and stability of the vehicle. In addition, by predicting the driving trajectory and extracting the front wheel turning angle, the vehicle's driving direction and posture can be controlled more accurately, improving the accuracy and reliability of remote takeover.
[0207] See also Figure 4 , Figure 4 This is a schematic diagram of the structure of a vehicle control component provided in an embodiment of the present application. Figure 4 As shown, the vehicle control component includes:
[0208] A remote driving console 401, a remote driving module 402 and an automatic driving module 403;
[0209] The remote driving station 401 is used to receive a dynamic takeover request of a target vehicle sent by a remote driver, wherein the driving mode of the target vehicle is an automatic driving mode;
[0210] The remote driving module 402 is used to process the dynamic takeover request of the target vehicle and forward the processed dynamic takeover request of the target vehicle to the automatic driving module;
[0211] The autonomous driving module 403 is used to query the dynamic takeover request of the target vehicle after information processing forwarded by the remote driving module;
[0212] In response to querying the dynamic takeover request of the target vehicle after information processing forwarded by the remote driving module, the automatic driving module 403 is further used to obtain the state model, observation model, initial state and sensor data of the target vehicle, and predict the driving trajectory of the target vehicle based on the state model, observation model, initial state and sensor data of the target vehicle;
[0213] The automatic driving module 403 is further used to extract the predicted value of the front wheel turning angle of the target vehicle from the driving trajectory;
[0214] The remote driving module 402 is further used to perform a steering angle calibration based on the predicted value of the front wheel steering angle to obtain a calibration result, and to remotely and dynamically take over the vehicle based on the calibration result.
[0215] In a possible implementation, the autonomous driving module 403 specifically includes:
[0216] The acquisition conversion module is used to obtain the predicted value of the front wheel turning angle of the target vehicle under the driving trajectory, and convert the predicted value of the front wheel turning angle of the target vehicle under the driving trajectory into a target steering wheel angle of the target vehicle.
[0217] In a possible implementation, the remote driving module 402 specifically includes:
[0218] A steering wheel angle mapping module, used for performing angle mapping on the target steering wheel angle to obtain a predicted mapping angle; the predicted mapping angle is a predicted value of a remote steering wheel angle of a remote driving console of the target vehicle;
[0219] An acquisition module, used for acquiring an adjusted remote steering wheel angle of the remote driving station and an adjusted vehicle steering wheel angle of the target vehicle; the adjusted remote steering wheel angle of the remote driving station is obtained by adjusting according to the predicted mapping angle; the adjusted vehicle steering wheel angle of the target vehicle is obtained by adjusting according to the target steering wheel angle;
[0220] A calculation module, used for obtaining an angle error by subtracting the adjusted remote steering wheel angle of the remote driving station from the adjusted vehicle steering wheel angle of the target vehicle;
[0221] The angle verification module is used to perform angle verification based on the angle error and the angle rule to obtain a verification result.
[0222] In a possible implementation, the remote driving module 402 further includes:
[0223] A state exit module, when the verification result is successful, is used to control the target vehicle to exit the automatic driving mode and perform remote dynamic takeover of the vehicle;
[0224] An execution module, when the verification result is in verification, is used to re-execute the state model, sensor data, observation model and initial state of the target vehicle to predict the driving trajectory of the target vehicle and subsequent steps;
[0225] The state maintaining module is used to control the target vehicle to maintain the automatic driving mode when the verification result is a verification failure, and to query the dynamic takeover request of the target vehicle again.
[0226] In a possible implementation, the corner rule includes:
[0227] When the angle error is less than the preset error, the verification result is fed back as verification success;
[0228] When the angle error is greater than or equal to the preset error, the verification result is fed back as being in verification;
[0229] When the verification result is fed back as being in verification, the steps of obtaining the state model, observation model, initial state and sensor data of the target vehicle and subsequent steps are executed again to obtain a new angle error. When the new angle error is still greater than or equal to the preset error within a preset time period, the verification result is changed to verification failure; the preset time period refers to the time period from obtaining the state model, observation model, initial state and sensor data of the target vehicle to a preset time; and / or,
[0230] When the verification result is fed back as verification in progress, the steps of obtaining the state model, observation model, initial state and sensor data of the target vehicle and subsequent steps are executed again to obtain a new angle error. When the new angle error is still greater than or equal to the preset error within a preset number of times, the verification result is changed to verification failure. The preset number of times is the number of times the steps of obtaining the state model, observation model, initial state and sensor data of the target vehicle and subsequent steps are executed again to obtain a new angle error.
[0231] In a possible implementation, the corner rule further includes:
[0232] During the angle verification process, the adjusted remote steering wheel angle of the remote driving station changes, and the verification result is fed back as verification failure; and / or,
[0233] During the turning angle verification process, the remote brake pedal of the target vehicle is subjected to force, and the verification result is fed back as verification failure; and / or,
[0234] During the turning angle verification process, a remote driving pedal of the target vehicle is subjected to force, and the verification result is fed back as verification failure.
[0235] In a possible implementation, the automatic driving module 402 further includes:
[0236] A position prediction module, used to predict the predicted position of the target vehicle at the next moment through an extended Kalman filter (EKF) algorithm based on the state model, observation model, initial state and sensor data of the target vehicle;
[0237] The driving trajectory prediction module is used to predict the driving trajectory of the target vehicle using a model predictive control (MPC) model based on the predicted position and trajectory constraints.
[0238] In a possible implementation, the trajectory constraint condition includes:
[0239] The front wheel steering angle is greater than or equal to -10 degrees and less than or equal to 10 degrees; and / or,
[0240] The front wheel steering angle changes by greater than or equal to -0.8 degrees and less than or equal to 0.8 degrees; and / or
[0241] The front wheel slip angle is greater than or equal to -2.8 degrees and less than or equal to 2.8 degrees; and / or,
[0242] The rear wheel slip angle is greater than or equal to -2.8 degrees and less than or equal to 2.8 degrees.
[0243] In a possible implementation, the vehicle control component further includes:
[0244] The prompt module is used to prompt the user that the target vehicle exits the automatic driving mode when the verification result is successful.
[0245] In addition, an embodiment of the present application further provides a processor for running a computer program, which, when running, executes the remote dynamic takeover method of the vehicle as described above.
[0246] The embodiment of the present application provides a vehicle control component, which first uses a remote driving console 401 to receive a dynamic takeover request of a target vehicle whose driving mode is an automatic driving mode sent by a remote driver. Then, the remote driving module 402 processes the dynamic takeover request of the target vehicle received by the remote driving console 401, and forwards the dynamic takeover request of the target vehicle after information processing to the automatic driving module 403. The automatic driving module 403 queries the dynamic takeover request of the target vehicle after information processing forwarded by the remote driving module 402. When the dynamic takeover request of the target vehicle after information processing forwarded by the remote driving module 402 is queried, the automatic driving module 403 will obtain the state model, observation model, initial state and sensor data of the target vehicle, and predict the driving trajectory of the target vehicle based on the state model, observation model, initial state and sensor data of the target vehicle. Then the automatic driving module 403 will continue to extract the predicted value of the steering wheel angle of the target vehicle under the driving trajectory, so that the remote driving module 402 can perform an angle calibration based on the predicted value of the front wheel angle to obtain a calibration result, and remotely dynamically take over the vehicle based on the calibration result. Compared with the method of requiring parking to unify angle control parameters, this application can achieve remote dynamic takeover of vehicles without parking. This can reduce the time cost in the takeover process and improve the overall operational efficiency. At the same time, by predicting the driving trajectory of the target vehicle and extracting the front wheel angle, it is possible to respond to the state changes of the target vehicle in a more real-time manner and make timely adjustments during the takeover process to ensure the safety and stability of the vehicle. In addition, by predicting the driving trajectory and extracting the front wheel angle, the driving direction and posture of the vehicle can be controlled more accurately, improving the accuracy and reliability of remote takeover.
[0247] The above is a detailed introduction to a remote dynamic takeover method for a vehicle, a vehicle control component and a processor provided by the present application. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the embodiments can be referenced to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of the present application.
[0248] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.
[0249] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.
[0250] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for remote dynamic takeover of a vehicle, characterized in that: Applied to a vehicle control assembly, the method comprises: Querying a dynamic takeover request of a target vehicle, wherein the driving mode of the target vehicle is an automatic driving mode; In response to a dynamic takeover request of the target vehicle, acquiring a state model, an observation model, an initial state, and sensor data of the target vehicle; Predicting a driving trajectory of the target vehicle based on a state model, sensor data, an observation model, and an initial state of the target vehicle; Extracting a predicted value of a front wheel turning angle of the target vehicle from the driving trajectory, and performing a turning angle calibration based on the predicted value of the front wheel turning angle to obtain a calibration result; Remote dynamic takeover of the vehicle is performed based on the verification result.
2. The method according to claim 1, characterized in that The step of extracting a predicted value of a front wheel turning angle of the target vehicle from the driving trajectory and performing a turning angle calibration based on the predicted value of the front wheel turning angle to obtain a calibration result includes: Acquire a predicted value of a front wheel turning angle of the target vehicle under the driving trajectory, and convert the predicted value of the front wheel turning angle of the target vehicle under the driving trajectory into a target steering wheel angle of the target vehicle; Performing angle mapping on the target steering wheel angle to obtain a predicted mapping angle; the predicted mapping angle is a predicted value of a remote steering wheel angle of a remote driving console of the target vehicle; Acquire an adjusted remote steering wheel angle of the remote driving station and an adjusted vehicle steering wheel angle of the target vehicle; the adjusted remote steering wheel angle of the remote driving station is obtained by adjusting according to the predicted mapping angle; the adjusted vehicle steering wheel angle of the target vehicle is obtained by adjusting according to the target steering wheel angle; Subtracting the adjusted remote steering wheel angle of the remote driving station from the adjusted vehicle steering wheel angle of the target vehicle to obtain an angle error; Angle verification is performed based on the angle error and the angle rule to obtain a verification result.
3. The method according to claim 1, characterized in that The remote dynamic takeover of the vehicle based on the verification result includes: When the verification result is successful, controlling the target vehicle to exit the automatic driving mode and remotely dynamically taking over the vehicle; When the verification result is in verification, the step of predicting the driving trajectory of the target vehicle and subsequent steps based on the state model, sensor data, observation model and initial state of the target vehicle is performed again; When the verification result is a verification failure, the target vehicle is controlled to maintain the automatic driving mode, and the dynamic takeover request of the target vehicle is queried again.
4. The method according to claim 2, characterized in that The corner rules include: When the angle error is less than the preset error, the verification result is fed back as verification success; When the angle error is greater than or equal to the preset error, the verification result is fed back as being in verification; When the verification result is fed back as being in verification, the steps of obtaining the state model, observation model, initial state and sensor data of the target vehicle and subsequent steps are executed again to obtain a new angle error. When the new angle error is still greater than or equal to the preset error within a preset time period, the verification result is changed to verification failure; the preset time period refers to the time period from obtaining the state model, observation model, initial state and sensor data of the target vehicle to a preset time; and / or, When the verification result is fed back as verification in progress, the steps of obtaining the state model, observation model, initial state and sensor data of the target vehicle and subsequent steps are executed again to obtain a new angle error. When the new angle error is still greater than or equal to the preset error within a preset number of times, the verification result is changed to verification failure. The preset number of times is the number of times the steps of obtaining the state model, observation model, initial state and sensor data of the target vehicle and subsequent steps are executed again to obtain a new angle error.
5. The method according to claim 4, characterized in that The corner rules also include: During the angle verification process, the adjusted remote steering wheel angle of the remote driving station changes, and the verification result is fed back as verification failure; and / or, During the turning angle verification process, the remote brake pedal of the target vehicle is subjected to force, and the verification result is fed back as verification failure; and / or, During the turning angle verification process, a remote driving pedal of the target vehicle is subjected to force, and the verification result is fed back as verification failure.
6. The method according to claim 1, characterized in that The predicting of the driving trajectory of the target vehicle based on the state model, observation model, initial state and sensor data of the target vehicle includes: Based on the state model, observation model, initial state and sensor data of the target vehicle, predict the predicted position of the target vehicle at the next moment by using an extended Kalman filter (EKF) algorithm; The driving trajectory of the target vehicle is predicted using a model predictive control (MPC) model based on the predicted position and trajectory constraints.
7. The method according to claim 6, characterized in that The trajectory constraints include: The front wheel steering angle is greater than or equal to -10 degrees and less than or equal to 10 degrees; and / or, The front wheel steering angle changes by greater than or equal to -0.8 degrees and less than or equal to 0.8 degrees; and / or The front wheel slip angle is greater than or equal to -2.8 degrees and less than or equal to 2.8 degrees; and / or, The rear wheel slip angle is greater than or equal to -2.8 degrees and less than or equal to 2.8 degrees.
8. The method according to claim 3, characterized in that The method further comprises: When the verification result is successful, the user is prompted that the target vehicle exits the automatic driving mode.
9. A vehicle control assembly, the vehicle control assembly comprising a remote driving console, a remote driving module and an automatic driving module; The remote driving station is used to receive a dynamic takeover request of a target vehicle sent by a remote driver; ,in, The driving mode of the target vehicle is an automatic driving mode. The remote driving module is used to process the dynamic takeover request of the target vehicle and forward the processed dynamic takeover request of the target vehicle to the automatic driving module; The autonomous driving module is used to query the dynamic takeover request of the target vehicle after information processing forwarded by the remote driving module; In response to querying the dynamic takeover request of the target vehicle after information processing forwarded by the remote driving module, the automatic driving module is further used to obtain the state model, observation model, initial state and sensor data of the target vehicle, and predict the driving trajectory of the target vehicle based on the state model, observation model, initial state and sensor data of the target vehicle; The automatic driving module further extracts a predicted value of a front wheel turning angle of the target vehicle from the driving trajectory; The remote driving module is further used to perform a turning angle calibration based on the predicted value of the front wheel turning angle to obtain a calibration result, and to perform remote dynamic takeover of the vehicle based on the calibration result.
10. A processor, characterized in that: Used to run a computer program, which, when running, executes the remote dynamic takeover method for a vehicle as described in any one of claims 1-8.
