CN111506054B - Automatic guiding method of 4WID-4WIS robot chassis - Google Patents

Abstract

The present invention relates to an automatic guidance method for a 4WID‑4WIS robot chassis, wherein the robot chassis is provided with four wheel groups, and the method comprises: 1. obtaining the current position of the robot; 2. obtaining a global path from the current position to the target position, obtaining the target point and the local path in the local area, and obtaining the motion parameters of the robot in the next period of time according to the local path; 3. establishing a rectangular coordinate system with the center of the robot chassis as the origin according to the motion parameters of the robot, and obtaining the control instructions of each of the four wheel groups; 4. obtaining the actual motion of the robot chassis by inverse kinematics solution, and adjusting the coordinated control process of the four wheel groups according to the actual motion of the four wheel groups. Compared with the prior art, the present invention realizes the coordinated control of the four wheel groups, avoids the adverse effects caused by the incompatibility of the control instructions with the wheel groups as much as possible, and improves the accuracy of robot control.

Description

An automatic guidance method for 4WID-4WIS robot chassis

Technical Field

The present invention relates to the field of robot technology, and in particular to an automatic guiding method of a 4WID-4WIS robot chassis.

Background Art

ROS (Robot Operating System) is a robot software platform. ROS can be divided into two layers. The lower layer is the operating system layer, and the higher layer is various software packages contributed by a large user base to achieve different functions, such as positioning mapping, action planning, perception, simulation, etc.

The low-level ROS uses the BSD license, so it is open source software and can be used for research and commercial purposes free of charge. The high-level user-provided packages can use many different licenses.

The focus of robot movement includes kinematic solutions for sending commands and inverse kinematic solutions for feedback. For traditional navigation algorithm models, the ultimate control goal is to treat the vehicle body as a point mass, and complete the issuance of commands by giving the linear velocity and angular velocity of this point mass along the three coordinate axes in three-dimensional space. For common two-wheel or omnidirectional structures, the body's spin, forward movement, steering and other processes can be seamlessly switched. At the same time, its response to angular velocity changes can be directly mapped to the motor speed control, thereby completing target control in near real time.

However, for the four-wheel independent drive-independent steering (4WID-4WIS) system, the process of changing the angular velocity includes the coordinated control of the steering motor and the drive motor. For many control scenarios, such as switching from execution to self-rotation, the steering motor needs to pre-set the direction of the drive motor before the drive operation can be performed. The above characteristics make the control of this system often procedural. How to connect the 4WID-4WIS robot to a mature real-time instruction system has become the biggest focus and difficulty.

There are relatively mature open source implementations of autonomous robot navigation. However, among these implementation methods, only the standard differential system and the design of the universal wheel or Mecanum wheel structure are targeted. There is a characteristic in all such designs, that is, the non-exclusiveness of the system's response to the target signal, which will become a difficulty in the implementation of the autonomous 4WIS-4DIS frame, because this structure requires the coordinated operation of multiple mechanisms, and even sequential forward and backward execution operations.

Summary of the invention

The purpose of the present invention is to provide an automatic guidance method for a 4WID-4WIS robot chassis in order to overcome the defects of the above-mentioned prior art.

The purpose of the present invention can be achieved by the following technical solutions:

An automatic guidance method for a 4WID-4WIS robot chassis, wherein the robot chassis is provided with four wheel sets, the method comprising:

S1. Get the current position of the robot;

S2, obtain the global path from the current position to the target position, divide the local area around the current position from the global path, obtain the target point and local path of the local area, and obtain the motion parameters of the robot in the next period of time according to the local path;

S3. According to the motion parameters of the robot, a rectangular coordinate system is established with the center of the robot chassis as the origin to obtain control instructions for the motion speed, motion direction and rotation angular velocity of each of the four wheel groups;

S4. In the process of controlling the movement of the robot according to the control instructions, the actual movement conditions of the four wheel groups are collected, the actual movement conditions of the robot chassis are obtained through inverse kinematics solution, and the coordinated control process of the four wheel groups is adjusted according to the actual movement conditions of the four wheel groups.

Preferably, the kinematic inverse solution includes linear velocity inverse solution and angular velocity inverse solution.

Preferably, the inverse solution of the linear velocity includes: taking two diagonal wheel groups of the four wheel groups as a group, and taking the sum of the velocity vectors of the two wheel groups in one group as the actual linear velocity of the robot chassis.

Preferably, the inverse solution of the angular velocity includes: collecting angular acceleration data of the robot chassis through an inertial navigation measurement unit, and then obtaining the angular velocity of the robot chassis through an integral operation.

Preferably, the process of adjusting the coordinated control of the four wheel groups according to the actual movement conditions of the four wheel groups includes: when the current wheel group will exceed the angle limit of the wheel group after executing the control instruction, the wheel group rotates in the opposite direction to its supplementary angle, and at the same time reverses the speed of the wheel group before executing the control instruction.

Preferably, the process of adjusting the coordinated control of the four wheel sets according to the actual movement conditions of the four wheel sets includes: when a maximum value v _max among the speeds of the four wheel sets is greater than a speed limit value v _limit , scaling the speeds of the four wheel sets by v _limit /v _max .

Preferably, the process of adjusting the coordinated control of the four wheel sets according to the actual movement conditions of the four wheel sets includes: when the target turning angle value of the steering operation of the wheel set is greater than the set turning angle limit value, judging whether the collected speed of the wheel set is greater than a predetermined threshold value v ₀ , and if so, braking the wheel set to reduce the speed to below the threshold value v ₀ before performing the steering operation.

Preferably, the process of adjusting the coordinated control of the four wheel sets according to the actual movement conditions of the four wheel sets includes: when the collected acceleration of the wheel set exceeds the set acceleration limit value, generating a control instruction to decompose the acceleration of the wheel set into control instructions that meet the acceleration limit value.

Preferably, the robot chassis includes a chassis frame, a controller and four wheel sets arranged on the chassis frame, the wheel set includes a protective frame and a drive motor, an encoder, a planetary reducer, a coupling and a wheel arranged in the protective frame and connected in sequence; the protective frame includes a horizontal upper baffle and a lower baffle, the ends of the upper baffle and the lower baffle are respectively connected to the axial part on the inner side of the wheel, and a first vertical plate is provided between the planetary reducer and the coupling; a steering transmission member is provided on the upper side of the upper baffle corresponding to the planetary reducer, and the steering transmission member is sleeved on the output shaft of the steering motor located above it, and the steering motor is arranged in a protective shell fixedly connected to the chassis frame; the encoders, drive motors and steering motors of the four wheel sets are respectively connected to the controller.

Preferably, a laser radar is provided on the upper side of the chassis frame, and the laser radar is connected to the controller.

Compared with the prior art, the present invention obtains control instructions for four wheel groups by solving the kinematics of the 4WID-4WIS robot chassis, obtains the comprehensive speed and angular velocity of the robot chassis by inversely solving the actual motion conditions of the four wheel groups, and realizes coordinated control of the four wheel groups, avoiding adverse effects caused by the incompatibility between the control instructions and the wheel groups as much as possible, ensuring that the issued control instructions are consistent with the posture orientation of the wheel groups at the moment of being controlled, and ensuring that the four wheel groups receive the control instructions to make the vehicle body move in the desired control direction without causing inconsistent speeds or contradictory angles between the wheels, thereby causing errors in the movement of the vehicle body, thereby improving the accuracy of robot control.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of a 4WID-4WIS robot chassis automatic guidance method flow chart;

FIG2 is a schematic diagram of a robot chassis in an embodiment;

FIG3 is a schematic diagram of the path before and after optimization in the embodiment;

FIG4 is a rear view schematic diagram of a robot chassis in an embodiment;

FIG. 5 is a schematic diagram of the structure of a wheel set in an embodiment.

Markings in the figure: 1. chassis frame, 2. controller, 3. wheel set, 4. laser radar, 5. drive motor, 6. encoder, 7. second wire outlet hole, 8. planetary reducer, 9. coupling, 10. wheel, 11. upper baffle, 12. lower baffle, 13. first vertical plate, 14. guard plate, 15. steering transmission part, 16. steering motor, 17. support plate, 18. second vertical plate, 19. upper shield plate, 20. first wire outlet hole, 21. connecting plate.

DETAILED DESCRIPTION

The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

Example

The present application proposes an automatic guidance method for a 4WID-4WIS robot chassis. The robot chassis is provided with four wheel sets, and the relatively mature ROS is selected as the navigation system platform.

The method includes:

S1. Get the current position of the robot;

The robot's positioning system obtains its own position estimation by integrating information from wheel encoders and detection devices such as lidar, etc., using the existing Monte Carlo algorithm;

S2. Obtain a global path from the current position to the target position through an existing path finding algorithm, divide a local area around the current position from the global path, and obtain the target point and local path of the local area through an existing simulation calculation method, and obtain the motion parameters of the robot in the next period of time according to the local path;

S3. According to the motion parameters of the robot, a rectangular coordinate system is established with the center of the robot chassis as the origin to obtain control instructions for the motion speed, motion direction and rotation angular velocity of each of the four wheel groups;

S4. In the process of controlling the movement of the robot according to the control instructions, the actual movement conditions of the four wheel groups are collected, the actual movement conditions of the robot chassis are obtained through inverse kinematics solution, and the coordinated control process of the four wheel groups is adjusted according to the actual movement conditions of the four wheel groups.

The kinematic solution and inverse solution of the four-wheel system are the basis for the entire robot chassis access control system. When omnidirectional control is enabled, for a robot moving in a plane, the direction perpendicular to the wheel axis is the X-axis direction, and the direction parallel to the wheel axis is the Y-axis direction. The instructions obtained should contain three parts: the linear velocity v _x in the X-axis direction, the linear velocity v _y in the Y-axis direction, and the rotation angular velocity ω. The three are synthesized to obtain the robot rotating along a certain point on the plane. In this embodiment, in order to simplify the description, a point outside the vehicle body is taken as the rotation point A for description, as shown in Figure 2.

For the rotational motion of the vehicle body around point A, that is, the vehicle body moves around it at a certain radius and linear velocity. The linear velocity of the vehicle body is determined by v _x and v _y . The directions of v _x and v _y are perpendicular to each other. According to the composite calculation of the velocity, we can get:

The direction of v _xy is also obtained from v _x and v _y . For the first quadrant, its angle relative to the X axis should be:

α＝arctan( _vy / _vx )

For the case where the rotation speed is not considered, the v _xy and angle α calculated here are the rotation angle and driving speed shared by the four wheel sets.

When angular velocity is added, the motion of the four wheels is equivalent to the synthesis of linear motion and pure rotational motion. Since the linear velocity _vxy and the angular velocity ω are known, the distance between the rotation center A and the center point O of the robot chassis, i.e., the rotation radius, is R = _vxy /ω.

The angular position of the rotation center A relative to point O can also be given by the direction of _vxy . Since R is always perpendicular to _vxy , according to the signs of the original velocities _vx and _vy , the direction of the rotation center A relative to point O can be comprehensively determined as _αA = α±π/2. The positive and negative sign of this formula is determined by the directions of _vx and _vy .

Therefore, this method takes the center of the robot chassis as the origin (0, 0) of the rectangular coordinate system, and can obtain the coordinates of the rotation center point A (Rcosα _A , Rsinα _A ).

In the speed control calculation, the robot chassis is simplified to a point mass without length and width. However, when it comes to actual wheel control, the position of the wheel must be considered. As shown in Figure 2, it can be clearly seen that the rotation angles of the four wheel groups will be significantly different due to the size of the robot chassis.

Assuming that the length and width of the robot chassis are H and D respectively, then in the rectangular coordinate system established at the center of the chassis O, the coordinates of the left front wheel group are (-D/2, H/2), and the right front wheel group is (D/2, H/2). Similarly, the coordinates of the left and right rear wheel groups are (-D/2, -H/2) and (D/2, -H/2) respectively.

In this way, after the rotation center (Rcosα _A , Rsinα _A ) and the coordinates of the four wheel sets are known, four vectors U1, U2, U3, and U4 pointing from the rotation center A to the centers of the four wheel sets can be obtained. The U1 vector is specifically (-D/2-Rcosα _A , H/2-Rsinα _A ), the U2 vector is (D/2-Rcosα _A , H/2-Rsinα _A ), the U3 vector is (-D/2-Rcosα _A , -H/2-Rsinα _A ), and the U4 vector is (D/2-Rcosα _A , -H/2-Rsinα _A ).

Take the coordinates of the four wheel group centers as the starting points of the new vectors, and make four new vectors V1, V2, V3, and V4 perpendicular to U1, U2, U3, and U4 respectively. The directions of the new vectors are the movement directions of the four wheel axles. V1 is (H/2-Rsinα _A , D/2+Rcosα _A ), V2 is (H/2-Rsinα _A , -D/2+Rcosα _A ), V3 is (-H/2-Rsinα _A , D/2+Rcosα _A ), and V4 is (-H/2-Rsinα _A , -D/2+Rcosα _A ).

The speeds of the four wheel axles are determined by the aforementioned U1~U4 vectors and the angular velocity ω. The speed of the left front wheel is |V1|＝|U1|·ω, the speed of the right front wheel is |V2|＝|U2|·ω, the speed of the left rear wheel is |V3|＝|U3|·ω, and the speed of the right rear wheel is |V4|＝|U4|·ω.

In this way, the movement direction and speed of the four wheel axles, that is, the direction and size of the vectors V1, V2, V3, and V4, can be obtained, completing the kinematic solution of the robot chassis control, and obtaining the control instructions for the movement speed, movement direction, and rotation angular velocity of each of the four wheel groups in step S3.

However, for the navigation system, the robot's feedback data is an important input source for self-positioning estimation. For the robot, the encoders and angle sensors installed on the four wheel drives obtain the real speed and direction of the four wheels, which need to be synthesized into the robot's comprehensive speed and angular velocity through an algorithm and fed back to the upper system. This process is called the inverse kinematic solution of the vehicle body. The inverse kinematic solution in step S4 includes the inverse linear velocity solution and the inverse angular velocity solution.

Compared with traditional models, the four-wheel independent drive system has a much more complicated inverse solution than the traditional control system. Taking the traditional two-wheel differential structure as an example, its input is the speed of the two wheels, and the position and direction of the two wheels are constant. The superposition effect can directly perform a comprehensive calculation of the two speeds, and obtain a point on the axle and extension line of the vehicle body as the center of rotation, so as to calculate the speed and angular velocity of the vehicle body. The whole process has a unique solution.

As for 4WID-4WIS, since there are bound to be errors in real-world control, it means that during the four-wheel active control process, the wheels will inevitably slip slightly. When the sampling speed is fast enough, the sampling data is small and the introduced error cannot be ignored.

For a four-wheel system, the four wheels are paired in pairs, with a total of 4x3/2=6 combinations. Each of these six combinations can uniquely determine a set of common speeds and angular velocities. This makes it difficult to solve the inverse motion. As the actual situation changes rapidly, it is difficult to complete data fusion in a relatively stable way.

According to the triangle stability principle, when a single wheel of a four-wheel vehicle fails due to changes in terrain, center of gravity, etc., the adjacent wheels and the wheels on the diagonal should be in contact with the ground. Therefore, this method only takes the diagonal wheel pairs, i.e. the left front and right rear, right front and left rear wheels as sampling pairs, and takes the average of the results as the estimated linear velocity of the robot chassis. The inverse solution of the linear velocity includes:

For one of the wheel pairs, when the two wheel speeds v ₁ , v ₂ and the steering angles d ₁ , d ₂ are known, its linear velocity is obtained by vector synthesis, and its estimated linear velocity in the X-axis and Y-axis directions can be obtained respectively:

ev _x =cosd ₁ ·v ₁ +cosd ₂ ·v ₂

ev _y =sind ₁ ·v ₁ +sind ₂ ·v ₂

According to the experimental results, the linear velocity obtained by the inverse kinematic solution is basically within the acceptable error range, while the error introduced by the angular velocity due to the small sampling value has caused serious difficulties in the calculation of this degree of freedom. This method decides to use an external sensor, the inertial measurement unit, to complete the collection of angular velocity, and use an encoder located at the motor output shaft on the wheel set to complete the collection of linear velocity. The inertial navigation sensor, also known as the inertial measurement unit (IMU), has a basic structure consisting of a three-axis accelerometer and a three-axis angular accelerometer, i.e., a gyroscope. Some inertial navigation systems can obtain the absolute direction of the heading angle through a geomagnetic sensor, but considering the influence of the driving circuit current in the robot chassis on the weak geomagnetic field and the uncertainty of the surrounding application environment, the geomagnetic sensor is not a reliable solution. Therefore, this method uses an accelerometer and an angular accelerometer for attitude solution.

The angular velocity of the wheel set is obtained by the inertial measurement unit. The inverse solution of the angular velocity includes: the angular acceleration data collected by the gyroscope can be integrated to obtain the angular velocity value.

However, due to the noise of the acquisition device itself and the uncertainty of the motion state, inputting the data into the system without any processing will inevitably introduce huge errors. Therefore, comprehensive acceleration and gyroscope data, filtering and mixing can greatly improve the error situation.

Kalman filtering of the raw data can perform linear optimization on the data and eliminate the noise introduced by the system circuit to a large extent. By collecting data in a static state for a period of time, the initial error of the system can be obtained and then eliminated. By processing the filtered data, the current vehicle head direction and instantaneous vehicle rotation speed data can be obtained.

After many experiments, the estimated angular velocity based on the inertial measurement unit has achieved good application results. Since the inertial measurement unit inevitably produces data drift, this is a cumulative error that cannot be eliminated as a relative quantity sensor. Therefore, when reading the heading angle, the movement of the comprehensive wheel group is used to read the difference in the estimated heading angle only when the robot is moving, which can avoid the accumulation of heading errors after long-term static state.

For the 4WID-4WIS robot, the biggest control difficulty lies in the coordinated control of the wheel groups. In this regard, this method has been optimized in many aspects. In the control driver, the results of the motion solution are tracked and post-processed in real time. The coordinated control of the four wheel groups is adjusted according to the actual motion conditions of the four wheel groups, mainly focusing on the following points:

1. Limitation: Since the steering structure of the 4WIS system cannot achieve continuous 360-degree rotation, it is necessary to judge the angle obtained by the algorithm. If the current wheel set will exceed the angle limit of the wheel set after executing the control command, the wheel set will rotate in the opposite direction to its supplementary angle, and at the same time reverse the speed of the wheel set before executing the control command.

2. Conversion: Due to the effect of the inner and outer wheel difference, the speed of each wheel group is different. The outer wheel group may plan a speed value that exceeds its execution capability. At this time, it is necessary to sample the speeds of all wheels to obtain the maximum value v _max of the four independent wheel group speeds. This speed is calculated together with the speed limit value v _limit . For the case of v _max ＞v _limit , the speeds of the four wheel groups need to be scaled by v _limit /v _max to ensure their normal operation.

3. Status judgment: Monitor the target steering angle values of the four wheel sets. When the target steering angle value of the wheel set steering operation is greater than the set steering angle limit value, such as 30°, the wheel set speed should be comprehensively considered to determine whether the collected wheel set speed is greater than the preset threshold value v _0. If so, the steering control command is first placed backwards, requiring the wheel set to brake first. When the speed drops below the threshold value v ₀ , steering and subsequent operations are performed to avoid severe slippage caused by sudden changes in speed at high speed.

4. Limit the speed of change, that is, acceleration limit: For speed changes exceeding the limit, generate control instructions to decompose the acceleration of the wheel set into control instructions that meet the acceleration limit to avoid slip errors caused by the vehicle body's inability to respond.

The planning parameters are adjusted as much as possible in the command planner. The optimized parameters mainly include the given speed upper and lower limits, acceleration upper and lower limits, and weighted values when each penalty condition is accumulated during route calculation. The planning process is actually to construct the objective function, multiply the various limit value parameters by the weights, and then perform comprehensive calculations to find the limit value of the objective function. The components of the objective function are the penalty values of each considered condition, including the straight-line distance from the target point, the accumulation of the distance from each obstacle on the path (including the wall), the nonlinearity of the vehicle body on a section of the route, etc. The farther the vehicle body is from the target, the closer the distance to the obstacle, and the higher the nonlinearity of the route, the greater the penalty score, and the route corresponding to the minimum value of the objective function is the optimal route. By adjusting the penalty value of the heading adjustment behavior, the response distance and response degree to the obstacle are adjusted to a certain extent, so that the final planned route tends to be a smooth straight line and a large-angle curve, thereby avoiding the adverse effects caused by the incompatibility between the command and the vehicle body as much as possible, as shown in Figure 3.

ROS is still a platform that is more oriented towards learning and research. In order to enable it to communicate easily with external devices and work stably and reliably, the robot has established a complete external communication interface, closing the system into a relatively independent part. In this way, when connecting upper commands with other devices, there is no need to directly access this system, which improves the stability and security of the system.

The robot's external communication is implemented using serial port communication. By defining a set of command response tables, it works in a question-and-answer manner. The navigation system host acts as the slave in the communication, receiving commands and providing feedback on the status.

Instructions mainly include command and query. Through commands, you can quickly switch pre-stored maps, set your own estimated position or navigation target position. The query command can obtain various information such as the robot's real-time position, direction, power, working status, etc. By cooperating with the upper device, the robot chassis can complete automated work, automatic charging, and even take the elevator to travel between different floors. By loading different maps, it can complete tasks that are difficult to complete under the native ROS navigation framework.

The robot chassis structure in this embodiment is shown in FIG4 , including a chassis frame 1 and a controller 2, a laser radar 4 and four wheel sets 3 arranged on the chassis frame 1. The laser radar 4 is connected to the controller 2, and specifically adopts the UST-10LX model of Hokuyo. In this embodiment, the controller 2 is specifically a computer.

As shown in Fig. 5, the wheel set 3 includes a protection frame and a driving motor 5, an encoder 6, a planetary reducer 8, a coupling 9 and a wheel 10 which are arranged in the protection frame and connected in sequence. The protection frame includes a horizontal upper baffle 11 and a lower baffle 12, and the ends of the upper baffle 11 and the lower baffle 12 are respectively connected to the inner axial part of the wheel 10. A first vertical plate 13 is provided between the planetary reducer 8 and the coupling 9 to increase the supporting capacity.

A steering transmission member 15 is provided on the upper side of the upper baffle 11 corresponding to the planetary reducer 8, and the steering transmission member 15 is sleeved on the output shaft of the steering motor 16 located above it. The steering motor 16 is arranged in a protective shell fixedly connected to the chassis frame 1, and the protective shell is connected to the bottom of the side of the chassis frame 1 through a connecting plate 21. In this embodiment, the steering transmission member 15 is specifically a rotating shaft fixed on the upper side of the upper baffle 11. The encoders 6, the drive motors 5 and the steering motors 16 of the four wheel sets 3 are respectively connected to the controller 2.

The lower side of the lower baffle plate 12 corresponding to the planetary reducer 8 is rotatably connected to a horizontally arranged support plate 17. The support plate 17 extends in the direction of the drive motor 5 and is connected to an upper shield plate 19 located above the protective frame through a second vertical plate 18. The end of the upper shield plate 19 can be movably sleeved on the outer side of the steering transmission member 15.

The second vertical plate 18 is fixedly connected to the chassis frame 1 and is provided with a first wire outlet hole 20 . The lower baffle plate 12 is provided with a second wire outlet hole 7 , which is beneficial to the wiring of the data line.

A guard plate 14 is provided above the top of the wheel 10 , and two sides of the guard plate 14 are respectively connected to the upper baffle 11 on the inner side of the wheel 10 and the axial part on the outer side of the wheel 10 . The guard plate 14 can reduce the impact of the environment on the wheel 10 .

Claims (6)

1. An automatic guidance method for a 4WID-4WIS robot chassis, wherein the robot chassis is provided with four wheel sets, characterized in that the method comprises: S1. Get the current position of the robot; S2, obtain the global path from the current position to the target position, divide the local area around the current position from the global path, obtain the target point and local path of the local area, and obtain the motion parameters of the robot in the next period of time according to the local path; S3. According to the motion parameters of the robot, a rectangular coordinate system is established with the center of the robot chassis as the origin to obtain control instructions for the motion speed, motion direction and rotation angular velocity of each of the four wheel groups; S4. In the process of controlling the movement of the robot according to the control instructions, the actual movement conditions of the four wheel groups are collected, the actual movement conditions of the robot chassis are obtained by inverse kinematics solution, and the coordinated control process of the four wheel groups is adjusted according to the actual movement conditions of the four wheel groups; The process of adjusting the coordinated control of the four wheel sets according to the actual movement conditions of the four wheel sets includes: If the current wheel set will exceed the wheel set angle limit after executing the control command, the wheel set will rotate in the opposite direction to its supplementary angle, and at the same time reverse the speed of the wheel set before executing the control command again; When the maximum value v _max among the speeds of the four wheel groups is greater than the speed limit value v _limit , the speeds of the four wheel groups are scaled by v _limit /v _max ; When the target turning angle value of the steering operation of the wheel set is greater than the set turning angle limit value, it is determined whether the collected wheel set speed is greater than the preset threshold value v ₀ . If so, the wheel set is braked to reduce the speed to below the threshold value v ₀ before the steering operation is performed. When the collected acceleration of the wheel set exceeds the set acceleration limit value, a control instruction is generated to decompose the acceleration of the wheel set into a control instruction that meets the acceleration limit value. 2. The automatic guidance method of a 4WID-4WIS robot chassis according to claim 1 is characterized in that the kinematic inverse solution includes linear velocity inverse solution and angular velocity inverse solution. 3. The automatic guidance method of a 4WID-4WIS robot chassis according to claim 2 is characterized in that the inverse solution of the linear velocity includes: taking two diagonal wheel groups of the four wheel groups as a group, and taking the velocity vector sum of the two wheel groups in one group as the actual linear velocity of the robot chassis. 4. The automatic guidance method of a 4WID-4WIS robot chassis according to claim 2 is characterized in that the inverse solution of the angular velocity includes: collecting angular acceleration data of the robot chassis through an inertial navigation measurement unit, and then obtaining the angular velocity of the robot chassis through an integral operation. 5. The automatic guidance method of a 4WID-4WIS robot chassis according to claim 1 is characterized in that the robot chassis includes a chassis frame, a controller and four wheel sets arranged on the chassis frame, the wheel set includes a protective frame and a drive motor, an encoder, a planetary reducer, a coupling and a wheel arranged in the protective frame and connected in sequence; the protective frame includes a horizontal upper baffle and a lower baffle, the ends of the upper baffle and the lower baffle are respectively connected to the axial part on the inner side of the wheel, and a first vertical plate is provided between the planetary reducer and the coupling; a steering transmission member is provided on the upper side of the upper baffle corresponding to the planetary reducer, and the steering transmission member is sleeved on the output shaft of the steering motor located above it, and the steering motor is arranged in a protective shell fixedly connected to the chassis frame; the encoders, drive motors and steering motors of the four wheel sets are respectively connected to the controller. 6. The automatic guidance method of a 4WID-4WIS robot chassis according to claim 5 is characterized in that a laser radar is provided on the upper side of the chassis frame, and the laser radar is connected to the controller.