Wheel mobile robot time delay teleoperation control system and method for safety obstacle avoidance

Abstract

A kind of wheel type mobile robot time delay teleoperation control system and method for security obstacle avoidance belong to intelligent vehicle remote teleoperation control technical field, solve the problems existing in real-time, stability and man-machine cooperation of traditional teleoperation system.The method of the present application comprises: by constructing the virtual potential field of remote environment, the future position and speed information of wheel type mobile robot are predicted by using potential field gradient constraint, to compensate the feedback time delay of slave end.The mobile robot not only can actively avoid obstacles and update control according to human intention, improve the obstacle avoidance performance and control continuity, realize the more effective cooperation of man-machine.Moreover, it provides the predictive force feedback based on environment potential field, drives operator to update the expected control intention of slave end with predictability, not only improves the performance of slave end mobile robot, but also reduces the burden of master end operator.The present application is suitable for teleoperation system for wheel type mobile robot.

Description

Technical Field

[0001] This application relates to the field of remote teleoperation control technology for intelligent vehicles, and in particular to time-delay remote operation control of wheeled mobile robots. Background Technology

[0002] Teleoperation systems for wheeled mobile robots, with their flexible mobility, have become a core solution for replacing humans in complex, unstructured, and remote scenarios. These systems typically employ a master-slave architecture, with the master deployed in a safe control area, where an operator inputs control intentions via a teleoperation handle. The slave, equipped with a wheeled mobile robot, is deployed at a remote work site. The master and slave communicate via a link to achieve bidirectional transmission of control commands and feedback information.

[0003] In teleoperation systems, force feedback from the slave end enhances human perception of the remote environment. However, communication latency between the master and slave ends causes feedback information to lag, affecting the system's real-time performance and stability. The feedback information received by the master end lags behind the actual state. Simultaneously, the control commands received by the slave end also lag behind the actual control commands. Therefore, compensation is needed for both aspects to improve the interaction accuracy between the master and slave ends and the system's response speed. Furthermore, slave-end wheeled mobile robots always passively follow the master operator, lacking proactive cooperation between the robot and the operator. Summary of the Invention

[0004] The purpose of this invention is to solve the technical problems of traditional teleoperation systems in terms of real-time performance, stability, and human-machine collaboration, and to provide a time-delay teleoperation control system and method for wheeled mobile robots with safety obstacle avoidance.

[0005] The technical solution adopted by the present invention to solve the above problems is: a time-delay remote operation control system for wheeled mobile robots for safe obstacle avoidance, the system comprising: master and slave computers, a communication module, slave force feedback compensation, master controller, master control compensation, and slave controller;

[0006] The master and slave computers are used to process the interaction and control information between the teleoperated handle and the slave wheeled mobile robot;

[0007] The communication module is used for real-time communication between the master computer and the slave computer, and there is a forward communication delay T1 and a reverse communication delay T2.

[0008] The master controller is used to calculate the desired control input of the master teleoperated handle;

[0009] The master control compensation is based on the environmental potential field constraint, predicts the future expected control input of the master handle, and compensates for the master control input.

[0010] The slave controller is used to calculate the desired speed of the slave wheeled mobile robot;

[0011] Force feedback compensation at the end is based on environmental potential field constraints, predicting the expected future speed of the wheeled mobile robot at the end, and compensating for the force feedback at the end.

[0012] Furthermore, the master-side control compensation includes a potential field constraint controller (PFC), which calculates the negative gradient of the environmental resultant potential field and introduces an environmental stiffness coefficient K. p Based on the equivalent mass M at the end, the virtual acceleration a of the wheeled mobile robot under the influence of the environmental resultant potential field is obtained. p :

[0013]

[0014] Where Δt is the control period. It is the gain coefficient. Here, t is the location of the autonomous vehicle, M is the discrete time, and U is the equivalent control quantity at the slave end. all It is the combined potential field, and ∇ is the gradient;

[0015] The master controller and slave controller include a predictor PFP based on environmental potential field constraints. The PFP predicts the future desired input from the master and the force feedback from the slave based on the vehicle motion model by introducing the environmental potential field and the control input from the master.

[0016] Furthermore, the master-side control compensation includes:

[0017] Let u r (t) represents the input r from the remote control master terminal of the wheeled mobile robot. m The desired velocity generated by matching under the action of (t-T1) is the desired velocity u of the wheeled mobile robot at time t under the guidance of the environmental potential field. s for:

[0018]

[0019] Slave status information v s (t) After a delay T2, the master receives v s (t−T2); Similarly, the master control input r m (t) After a delay T1, the slave receives r m (t−T1);

[0020] Define the state vector of the wheeled mobile robot between t and t+T2 as:

[0021]

[0022] Control input matrix for a wheeled mobile robot between t and T2:

[0023]

[0024] We can obtain:

[0025]

[0026] in, The initial state expansion matrix;

[0027] ,

[0028] in,

[0029]

[0030]

[0031] F p It is the matrix of forces acting on the environmental potential field.

[0032]

[0033] The control input matrix is ​​then:

[0034]

[0035] in, It is a time-weighted matrix.

[0036]

[0037] Combining the above formulas, the expected state matrix and expected velocity matrix of the wheeled mobile robot between t and t+T2 are obtained as follows:

[0038]

[0039] .

[0040] Furthermore, the master-side control compensation also includes:

[0041] The expected state and expected velocity of the master terminal relative to the slave terminal during the time interval from t+T2 to t+T2+T1 are calculated as follows:

[0042]

[0043] .

[0044] Furthermore, end force feedback compensation includes:

[0045] The state-space equation of a wheeled mobile robot in discrete time is defined as follows:

[0046]

[0047] Where A is the state transition matrix, B is the input matrix, and u is the control input;

[0048] The expected state and expected velocity of the wheeled mobile robot at time t+1 under the guidance of the environmental potential field are:

[0049]

[0050] Assuming that in the time interval from t to T2, the teleoperation r m Desired speed u acting on a wheeled mobile robot r Equal to the expected speed u of the wheeled mobile robot at the previous moment s ,Right now

[0051]

[0052] Where N represents the future prediction time domain;

[0053] will u s Substituting (t) into the state equation and expanding further, we get:

[0054]

[0055] Where ∇U is the derivative of the potential field function;

[0056] The system state over the next T time points is calculated using a recursive formula:

[0057]

[0058] In x(t+T), the first term represents the influence of the system's free evolution; the second term reflects the cumulative influence of the master-end control input; and the third term embodies the guiding role of the environmental potential field on the future state of the system; u s The first term in (t+T) reflects the wheeled mobile robot's position at time t, r. m The expected velocity generated under the action of (t-T1) is reflected in the second term, which reflects the cumulative effect of the environmental potential field on the guidance of the remote operating system.

[0059] Furthermore, the slave controller includes:

[0060] At time t, the master end will The master robot expects its position information to propagate forward to the slave wheeled robot at any given time; the slave wheeled robot actually receives... Information is based on feedback from the end. s (t-T1-T2) and the environmental potential field are jointly generated, based on the environmental potential field from v s (t-T1-T2) and the forward prediction of the effect of the environmental potential field. The expected information, among which, It is a prediction instruction from the slave end to the master end. It is the desired control of the environmental potential field;

[0061] Will The expected information and the actual speed information of the wheeled mobile robot are integrated and compensated to obtain the expected control quantity u from the master end of the teleoperation system acting on the slave end of the wheeled mobile robot. r for:

[0062]

[0063] in, It is a proportionality coefficient. It is the integral coefficient. It is a prediction instruction from the slave end to the master end. These are the differential coefficients.

[0064] Furthermore, the master controller includes:

[0065] From time t, the current vehicle speed v s (t), received from the master end r m (t-T1) information, based on the environmental potential field, predicts the expected future velocity u from the endpoint. s (t:t+T2), and combine it with the P control law. Together, these are fed back to the master terminal, which then receives the expected speed information U from the wheeled mobile robot after compensation. s (t-T2:t) information and force feedback information For u s The information (t-T2:t) and the actual information rm(t-T2:t) are used to perform error integral compensation on the current feedback, and the PID control law is used to calculate the feedback force at the slave end:

[0066]

[0067] Among them, C m It is a proportionality coefficient. It is the integral coefficient. It is the predicted state of the master-side autonomous vehicle to the slave-side vehicle. It is the differential coefficient; feedback force This reflects the tracking performance of the end-wheeled mobile robot under the combined influence of the remote control handle and the environmental potential field, reflecting the actual speed of the robot.

[0068] Furthermore, the real part of the main-terminal impedance of the system is greater than or equal to C. m The impedance at the slave end must be greater than C. s The absolute stability of the system is determined by K. m K s C m Cs The actual conditions determine that all are positive.

[0069] Secondly, the present invention provides a time-delayed teleoperation control method for a wheeled mobile robot for safe obstacle avoidance, the method comprising:

[0070] Calculate the desired control input of the master teleoperated handle;

[0071] Based on environmental potential field constraints, predict the future desired control input of the master handle and compensate for the master control input.

[0072] Calculate the desired speed of the end-wheeled mobile robot;

[0073] Based on environmental potential field constraints, the expected future speed of a wheeled mobile robot is predicted and the force feedback from the end is compensated.

[0074] Thirdly, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, it executes a time-delayed teleoperation control method for a wheeled mobile robot for safety obstacle avoidance as described above.

[0075] The beneficial effects of this invention are:

[0076] This invention constructs a virtual potential field in a remote environment and uses gradient constraints to predict the future position and velocity information of a wheeled mobile robot, thus compensating for feedback latency from the slave end. The mobile robot can not only proactively avoid obstacles and update control according to human intentions, improving obstacle avoidance performance and control continuity, but also achieve more effective human-machine collaboration. Furthermore, it provides predictive force feedback based on the environmental potential field, driving the operator to predictably update the slave end's desired control intentions, thereby improving the performance of the slave mobile robot and reducing the burden on the master operator.

[0077] This invention considers the impact of master-slave communication delay on the control system. By constructing an environmental potential field constraint for the wheeled mobile robot, it compensates for the influence of master-slave communication delay on slave force feedback and master control. Furthermore, based on the prediction of gradient constraint of the environmental potential field of the wheeled mobile robot, a teleoperation bilateral PID control law is designed, which greatly improves the accuracy and stability of teleoperation control under time delay conditions.

[0078] This invention is applicable to teleoperation systems for wheeled mobile robots. Attached Figure Description

[0079] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0080] Figure 1 This is a block diagram of a time-delayed remote operation control system for a wheeled mobile robot designed for safe obstacle avoidance, as described in this invention.

[0081] Figure 2 This is a system flowchart of a time-delayed remote operation control system for a wheeled mobile robot designed for safe obstacle avoidance, as described in this invention.

[0082] Figure 3 This is a time-delayed remote operation dual-port network diagram of a wheeled mobile robot remote operation control system for safe obstacle avoidance, as described in this invention.

[0083] Figure 4 This is a block diagram of a two-sided teleoperation system for a wheeled mobile robot based on a PID control law, which is described in this invention for a time-delay teleoperation control system for a wheeled mobile robot designed for safe obstacle avoidance. Detailed Implementation

[0084] Specific implementation method one: Combining Figure 1 This invention describes a time-delay teleoperation control system for a wheeled mobile robot designed for safe obstacle avoidance, relating to the field of remote teleoperation control technology for intelligent vehicles. The control system includes a master-slave computer 1, a communication module 2, a slave force feedback compensation module 3, a master controller 4, a master control compensation module 5, and a slave controller 6.

[0085] The master-slave computer 1 is mainly used to process the interaction and control information between the teleoperated handle and the slave wheeled mobile robot. The communication module 2 is mainly used for real-time communication between the master and slave computers, with a forward communication delay T1 and a reverse communication delay T2. The slave force feedback compensation 3 is mainly based on environmental potential field constraints to predict the future expected speed of the slave wheeled mobile robot and compensate for the slave force feedback. The master controller 4 is mainly used to calculate the expected control input of the master teleoperated handle. The master control compensation 5 is mainly based on environmental potential field constraints to predict the future expected control input of the master handle and compensate for the master control input. The slave controller 6 is mainly used to calculate the expected speed of the slave wheeled mobile robot.

[0086] Specific Implementation Method Two: Combining Figure 2 This invention describes an embodiment of the invention, which further defines the remote operation control system described above.

[0087] To establish a remote operation control system, a remote operating system control method is adopted that maps the position information of the master controller to the speed information of the slave vehicle; that is, the speed of the slave vehicle is mapped to the joint of the controller. A vehicle bilateral continuous remote operating system with time delay includes a human termination (HT), an environment termination (ET), a communication channel (CC), a master controller unit (MCU), and a slave controller unit (SCU). Among these, r... m The control input for master-side remote operation, u s For speed commands applied to a slave-end wheeled mobile robot, v s This represents the current speed of the slave device. The reaction forces between the slave and the environment. PFP is the environmental potential field gradient constraint predictor. PFC is the controller based on the environmental potential field gradient constraint. T1 and T2 are the communication delays between the master and slave ends.

[0088] Virtual potential fields are a commonly used algorithm for robot path planning and obstacle avoidance, especially suitable for dynamic and complex environments. This paper introduces the guiding role of the environmental virtual potential field to compensate for communication latency between the master and slave ends and improve interaction capabilities.

[0089] Define the position of the wheeled mobile robot as P. ego The obstacle is located at point P. obs The target point is located at P. tar If a global desired path exists, the target point can be obtained through the forward-looking global path, or it can be set by the operator on a remote terminal according to the local environment.

[0090]

[0091]

[0092]

[0093] Among them, U re It is the obstacle potential field, U at The gravitational potential field at the target point, U all It is a net potential field. The repulsive potential field U i re The attractive potential field U is determined by the repulsive force exerted by obstacle i on the wheeled mobile robot. att The gain coefficient K is determined by the attractive force exerted by the target point on the robot. re and K atIt is responsible for adjusting the strength of the potential field. The teleoperated master-slave controller will perform compensation control under the guidance of the gradient ∇Uall.

[0094] We calculate the negative gradient of the environmental resultant potential field and introduce the environmental stiffness coefficient K. p From the equivalent mass M at the end, we can obtain the virtual acceleration a of the wheeled mobile robot under the influence of the environmental resultant potential field. p .

[0095]

[0096] Where Δt is the control period.

[0097] Let u r (t) represents the input r from the remote control master terminal of the wheeled mobile robot. m The desired velocity generated by matching under the action of (t-T1) is the desired velocity u of the wheeled mobile robot at time t under the guidance of the environmental potential field. s for:

[0098]

[0099] The state-space equation of a wheeled mobile robot in discrete time is defined as follows:

[0100]

[0101] Where A is the state transition matrix, B is the input matrix, and u is the control input.

[0102] The expected state and expected velocity of the wheeled mobile robot at time t+1 under the guidance of the environmental potential field are:

[0103]

[0104] Since we cannot know the expected speed u of the wheeled mobile robot under the control input from the master end in the future within the prediction time from t to t+T2, r We can assume that in the time interval from t to T2, the teleoperation r m Desired speed u acting on a wheeled mobile robot r Equal to the expected speed u of the wheeled mobile robot at the previous moment s ,Right now

[0105]

[0106] This avoids relying on future, unknowable master-end control input information and also aligns with the potential field's guiding effect. (U) s Substituting (t) into the state equation and expanding further, we get:

[0107]

[0108] Considering the prediction time T≫Δt, the system state over the next T time intervals can be calculated using a recursive formula:

[0109]

[0110] In x(t+T), the first term represents the influence of the system's free evolution. The second term reflects the cumulative influence of the master-end control input. The third term reflects the guiding role of the environmental potential field on the future state of the system. s The first term in (t+T) reflects the wheeled mobile robot's position at time t, r. m The expected velocity generated under the action of (t-T1) is reflected in the second term, which reflects the cumulative effect of the environmental potential field on the guidance of the remote operating system.

[0111] Specific implementation method three: The implementation method of the present invention is a further limitation on the slave-end force feedback compensation 3 and master-end control compensation 5 as described above.

[0112] Among them, the force feedback compensation 3 at the end includes: predicting the reverse communication delay of the wheeled mobile robot at the end based on environmental potential field constraints. The expected future speed is calculated and the predicted expected speed information is fed back to the master end to compensate for the force feedback delay of the slave end.

[0113] Master control compensation 5 includes: predicting the forward communication delay of the master handle based on environmental potential field constraints. The system calculates the expected future control inputs and compensates for the master control inputs to reduce control lag caused by communication delays.

[0114] Due to communication delay, the slave status information v s (t) After a delay T2, the master receives v s (t−T2); Similarly, the master control input r m (t) After a time delay T1, the slave receives r m (t−T1). To perform compensation, we need to predict v. s (t:t+T2) and r m Information between (t:t+T1). Compared to information transmission at a single time t, this method contains more expected future information, which can effectively compensate for time delay errors.

[0115] We need to predict the expected speed of the wheeled mobile robot from time t to t+T2 and feed it back to the master. Between t and t+T2, the wheeled mobile robot may experience multiple discrete states (i.e., T2≫Δt), so we use matrix form to represent the state of the wheeled mobile robot.

[0116] Define the state vector of the wheeled mobile robot between t and t+T2 as:

[0117]

[0118] Control input matrix for a wheeled mobile robot between t and T2:

[0119]

[0120] We can obtain:

[0121]

[0122] in, This is the extended matrix of the initial state.

[0123] ,

[0124] in,

[0125]

[0126]

[0127] F p It is the matrix of forces acting on the environmental potential field.

[0128]

[0129] The control input matrix can then be rewritten as:

[0130]

[0131] in, It is a time-weighted matrix.

[0132]

[0133] Combining the above formulas, the expected state matrix and expected velocity matrix of the wheeled mobile robot between t and t+T2 are obtained as follows:

[0134]

[0135]

[0136] Finally, we need the expected future performance of the wheeled mobile robot guided by the potential field at time t. and the vehicle's actual speed v s (t) Feedback is sent to the master.

[0137] Similarly, we can obtain the expected state and expected velocity of the master end relative to the slave end during the time intervals from t+T2 to t+T2+T1 as follows:

[0138]

[0139] .

[0140] Specific implementation method four: Combination Figure 3 This embodiment is described in detail, and it further defines the master controller 4 and slave controller 6 as described above.

[0141] The slave controller 6 includes: At any given time, the slave controller 6 receives the master control input after the forward communication delay. And combined with the actual speed of the end-wheeled mobile robot and environmental potential field control quantity Calculate the desired control quantity of the master teleoperator acting on the slave wheeled mobile robot. Furthermore, the remote control quantity With environmental potential field control quantity The combined action of the two forces on the slave wheeled mobile robot yields the desired slave speed. .

[0142] The master controller 4 includes: slave devices in Get the current vehicle speed at all times It also receives control input from the master end after the forward communication delay. Predicting the expected future velocity from the edge based on the environmental potential field. The desired speed information and the proportional feedback term are then fed back to the master terminal. The master terminal controller 4 receives the compensated desired speed information. And after force feedback information, for With actual master input The error between them is compensated by integration, and the feedback force acting on the master remote control handle is calculated using a PID control law.

[0143] For a two-way teleoperation system for a wheeled mobile robot with timing delay, the two-port network within the dashed box can be described by the following formula:

[0144]

[0145] in, It is the interactive force applied by the operator to the main robot. Z is the interaction force between the end-user robot and its environment. ij These are elements of the impedance transfer matrix, used to characterize the dynamic mapping relationship between the master-slave end motion variables and the interaction force.

[0146] Remote control handle control quantity ur and environmental potential field control quantity u p The combined effect of u s For a slave-end wheeled mobile robot. The actual speed v of the slave-end wheeled mobile robot. s For the desired velocity u s The errors, together with the feedback, act on the equivalent torque of the main handle motor. That is, the feedback force felt by the master handle is, to some extent, a reflection of the tracking performance of the slave wheeled mobile robot on the desired speed, based on the actual speed.

[0147] As can be seen from the above, at time t, the master end will... The master robot expects its position information to propagate forward to the slave wheeled robot at any given time. The slave wheeled robot actually receives... Information is based on feedback from the end. s (t-T1-T2) and the environmental potential field are jointly generated. We rely on the environmental potential field from v s (t-T1-T2) and the forward prediction of the effect of the environmental potential field. The expected information. We will The expected information is integrated with the actual speed information of the wheeled mobile robot. This yields the expected control quantity u from the master teleoperator acting on the slave wheeled mobile robot. r for

[0148]

[0149] Where u s The combined effect of the remote control handle and the environmental potential field on the slave wheeled mobile robot is...

[0150]

[0151] As can be seen from the above, from time t, the current vehicle speed v s (t), received from the master end r m (t-T1) information, based on the environmental potential field, predicts the expected future velocity u from the endpoint. s (t:t+T2), and combine it with the P control law. Together, these are fed back to the master terminal, which then receives the expected speed information U from the wheeled mobile robot after compensation. s (t-T2:t) information and force feedback information For u s The information (t-T2:t) and the actual information rm(t-T2:t) are used to perform error integral compensation for the current feedback, and the PID control law is used to calculate the feedback force at the slave end.

[0152]

[0153] Among them, feedback force This reflects the tracking performance of the end-wheeled mobile robot under the combined influence of the remote control handle and the environmental potential field, reflecting the actual speed of the robot.

[0154] In a typical bidirectional teleoperation system, the P-control law is used to track the positions of the master and slave ends, while the D-control law is used to consume the reactive energy generated in the master-slave subsystem to maintain the passivity of the entire system. However, in the bidirectional teleoperation system for wheeled mobile robots proposed in this paper, the state variable r of the master robot... m (approximately equal to the position variable of the master robot) and the state variable v of the slave robot s To achieve matching, the P control law performs proportional control between the master and slave ends, reducing errors. The I control law is an integral compensation term for the time delay based on the predicted quantity guided by the environmental potential field, which can compensate for steady-state errors. Simultaneously, a D control law is introduced to consume the energy generated in each subsystem to maintain the passivity of the two-port network.

[0155] Specific Implementation Method Five: Combining Figure 4 This embodiment is described in detail, and it further defines the remote operation control system as described above.

[0156] The remote control employs a P+I+D control strategy, which, compared to the previous P+D controller, adds an integral compensation term, which helps compensate for the system's steady-state error. A two-port network model is established based on the designed control law, and stability analysis is performed. Applying a Laplace transform to the integral compensation term (initial conditions are 0), we obtain:

[0157]

[0158] Depend on Therefore, the two-port network model can be described as follows:

[0159]

[0160] To maintain the stability of a two-port network, according to the Llewellyn criterion, its control law parameters should satisfy:

[0161]

[0162] in, f is the control frequency; under high-frequency operation,

[0163] , ,

[0164] Therefore, the above formula can be simplified to:

[0165] Solving the above equation, we can obtain the absolute stability condition for a two-port network as follows:

[0166]

[0167] From the above equation, it can be seen that the real part of the main terminal impedance must be greater than or equal to C. m This ensures that the master-side system does not produce unstable oscillations at any frequency. The slave-side impedance must be greater than C. s This ensures that the slave device does not exhibit excessive errors when exposed to the environment. The absolute stability of the system is determined by K. m K s C m C s The real part of the condition is determined. Under normal operating frequency, the integral control term I... m I s The impact is relatively small, therefore its effect on the most basic stability condition can be ignored. Integral control term I m I s The effect of I approaches zero at high frequencies, therefore the final stability condition does not include I. m I s At the same time, these parameters must be positive to ensure the physical rationality of the system and avoid instability caused by negative damping. Therefore, by simply meeting these conditions in the design of the teleoperation controller, we can guarantee the stability of the system.

Claims

1. A time-delayed teleoperation control system for a wheeled mobile robot designed for safe obstacle avoidance, characterized in that: The system includes: master and slave computers, a communication module, slave force feedback compensation, master controller, master control compensation, and slave controller. The master and slave computers are used to process the interaction and control information between the teleoperated handle and the slave wheeled mobile robot; The communication module is used for real-time communication between the master computer and the slave computer, and there is a forward communication delay T1 and a reverse communication delay T2. The master controller is used to calculate the desired control input of the master teleoperated handle; The master control compensation is based on the environmental potential field constraint, predicts the future expected control input of the master handle, and compensates for the master control input. The slave controller is used to calculate the desired speed of the slave wheeled mobile robot; Force feedback compensation at the end is based on environmental potential field constraints, predicting the expected future speed of the wheeled mobile robot at the end, and compensating for the force feedback at the end.

2. The time-delay teleoperation control system for a wheeled mobile robot oriented towards safety obstacle avoidance according to claim 1, characterized in that: The master-side control compensation includes a potential field constraint controller (PFC), which calculates the negative gradient of the environmental resultant potential field and introduces an environmental stiffness coefficient K. p Based on the equivalent mass M at the end, the virtual acceleration a of the wheeled mobile robot under the influence of the environmental resultant potential field is obtained. p : Where Δt is the control period. It is the gain coefficient. Here, t is the location of the autonomous vehicle, M is the discrete time, and U is the equivalent control quantity at the slave end. all It is the combined potential field, and ∇ is the gradient; The master controller and slave controller include a predictor PFP based on environmental potential field constraints. The PFP predicts the future desired input from the master and the force feedback from the slave based on the vehicle motion model by introducing the environmental potential field and the control input from the master.

3. The time-delay teleoperation control system for a wheeled mobile robot oriented towards safety obstacle avoidance according to claim 1, characterized in that: Master-side control compensation includes: Let u r (t) represents the input r from the remote control master terminal of the wheeled mobile robot. m The desired velocity generated by matching under the action of (t-T1) is the desired velocity u of the wheeled mobile robot at time t under the guidance of the environmental potential field. s for: Slave status information v s (t) After a delay T2, the master receives v s (t−T2); Similarly, the master control input r m (t) After a delay T1, the slave receives r m (t−T1); Define the state vector of the wheeled mobile robot between t and t+T2 as: Control input matrix for a wheeled mobile robot between t and T2: We can obtain: in, The initial state expansion matrix; ， in, F p It is the matrix of forces acting on the environmental potential field. The control input matrix is ​​then: in, It is a time-weighted matrix. Combining the above formulas, the expected state matrix and expected velocity matrix of the wheeled mobile robot between t and t+T2 are obtained as follows: 。 4. A time-delay teleoperation control system for a wheeled mobile robot oriented towards safety obstacle avoidance, as described in claim 3, is characterized in that: Master-side control compensation also includes: The expected state and expected velocity of the master terminal relative to the slave terminal during the time interval from t+T2 to t+T2+T1 are calculated as follows: 。 5. A time-delay teleoperation control system for a wheeled mobile robot oriented towards safety obstacle avoidance, as described in claim 1, is characterized in that: Force feedback compensation from the end includes: The state-space equation of a wheeled mobile robot in discrete time is defined as follows: Where A is the state transition matrix, B is the input matrix, and u is the control input; The expected state and expected velocity of the wheeled mobile robot at time t+1 under the guidance of the environmental potential field are: Assuming that in the time interval from t to T2, the teleoperation r m Desired speed u acting on a wheeled mobile robot r Equal to the expected speed u of the wheeled mobile robot at the previous moment s ,Right now Where N represents the future prediction time domain; will u s Substituting (t) into the state equation and expanding further, we get: Where ∇U is the derivative of the potential field function; The system state over the next T time points is calculated using a recursive formula: In x(t+T), the first term represents the influence of the system's free evolution; the second term reflects the cumulative influence of the master-end control input; and the third term embodies the guiding role of the environmental potential field on the future state of the system; u s The first term in (t+T) reflects the wheeled mobile robot's position at time t, r. m The expected velocity generated under the action of (t-T1) is reflected in the second term, which reflects the cumulative effect of the environmental potential field on the guidance of the remote operating system.

6. A time-delay teleoperation control system for a wheeled mobile robot oriented towards safety obstacle avoidance, as described in claim 1, is characterized in that: The slave controller includes: At time t, the master end will The master robot expects its position information to propagate forward to the slave wheeled robot at any given time; the slave wheeled robot actually receives... Information is based on feedback from the end. s (t-T1-T2) and the environmental potential field are jointly generated, based on the environmental potential field from v s (t-T1-T2) and the forward prediction of the effect of the environmental potential field. The expected information, among which, It is a prediction instruction from the slave end to the master end. It is the desired control of the environmental potential field; Will The expected information and the actual speed information of the wheeled mobile robot are integrated and compensated to obtain the expected control quantity u from the master end of the teleoperation system acting on the slave end of the wheeled mobile robot. r for: in, It is a proportionality coefficient. It is the integral coefficient. It is a prediction instruction from the slave end to the master end. These are the differential coefficients.

7. A time-delay teleoperation control system for a wheeled mobile robot oriented towards safety obstacle avoidance according to claim 1, characterized in that: The master controller includes: From time t, the current vehicle speed v s (t), received from the master end r m (t-T1) information, based on the environmental potential field, predicts the expected future velocity u from the endpoint. s (t:t+T2), and combine it with the P control law. Together, these are fed back to the master terminal, which then receives the expected speed information U from the wheeled mobile robot after compensation. s (t-T2:t) information and force feedback information For u s The information (t-T2:t) and the actual information rm(t-T2:t) are used to perform error integral compensation on the current feedback, and the PID control law is used to calculate the feedback force at the slave end: Among them, C m It is a proportionality coefficient. It is the integral coefficient. It is the predicted state of the master-side autonomous vehicle to the slave-side vehicle. It is the differential coefficient; feedback force This reflects the tracking performance of the end-wheeled mobile robot under the combined influence of the remote control handle and the environmental potential field, reflecting the actual speed of the robot.

8. A time-delay teleoperation control system for a wheeled mobile robot oriented towards safety obstacle avoidance according to claim 1, characterized in that: The real part of the main-terminal impedance of the system is greater than or equal to C. m The impedance at the slave end must be greater than C. s The absolute stability of the system is determined by K. m K s C m C s The actual conditions determine that all are positive.

9. A time-delayed teleoperation control method for a wheeled mobile robot for safe obstacle avoidance, characterized in that: The method includes: Calculate the desired control input of the master teleoperated handle; Based on environmental potential field constraints, predict the future desired control input of the master handle and compensate for the master control input. Calculate the desired speed of the end-wheeled mobile robot; Based on environmental potential field constraints, the expected future speed of a wheeled mobile robot is predicted and the force feedback from the end is compensated.

10. A computer device, characterized in that: It includes a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, it executes the time-delayed teleoperation control method for a wheeled mobile robot for safety obstacle avoidance as described in claim 9.

Images