CN106882080B - Differential steering system and adaptive neural network fault-tolerant control method thereof - Google Patents

Abstract

The invention discloses a differential steering system and a fault-tolerant control method of a self-adaptive neural network thereof. In the running process, the whole vehicle electronic control unit acquires steering wheel rotation angle, yaw rate and vehicle speed signals in real time, calculates the difference value between the ideal yaw rate and the actual yaw rate, recalculates the output torque of the hub motor through the designed self-adaptive neural network controller, transmits the torque signals to the motor controller, and sends current signals to four hub motors through the motor controller to complete steering stability control under the normal and failure conditions of the hub motors. The invention can improve the reliability of the differential steering system and the stability and safety of the automobile during running.

Description

A differential steering system and its adaptive neural network fault-tolerant control method

Technical Field

The invention relates to the technical field of four-wheel steering, and in particular to a differential steering system and an adaptive neural network fault-tolerant control method thereof.

Background Art

For traditional vehicles, clutches, transmissions, drive shafts, differentials and even transfer cases are essential, and these components are not only heavy and complex in structure, but also require regular maintenance and have a high failure rate. However, wheel hub motors solve this problem very well. In addition to a simpler structure, vehicles driven by wheel hub motors can achieve better space utilization and high transmission efficiency.

Since the wheel hub motor has the characteristic of driving a single wheel independently, it is easy to realize front-wheel drive, rear-wheel drive or four-wheel drive. At the same time, the wheel hub motor can adjust the torque or speed of the left and right wheels to achieve differential steering, greatly reducing the turning radius of the vehicle, and in special circumstances, it can almost achieve on-site steering.

However, the hub motor may fail and have reliability issues. How to ensure the stability of the car in the event of a hub motor failure needs to be solved urgently.

Summary of the invention

The technical problem to be solved by the present invention is to provide a differential steering system and an adaptive neural network fault-tolerant control method thereof in view of the defects involved in the background technology.

The present invention adopts the following technical solutions to solve the above technical problems:

A differential steering system, comprising a steering wheel angle sensor, a steering wheel, a steering column, a rack and pinion steering gear, first to fourth wheels, first to fourth wheel hub motors, a front axle, a vehicle electronic control unit, a battery pack, a vehicle speed sensor, a yaw rate sensor, a rear axle and a motor control unit;

One end of the steering column is fixedly connected to the steering wheel, and the other end is connected to the front axle through a rack and pinion steering gear;

The steering wheel angle sensor is arranged on the steering column and is used to obtain the steering wheel angle;

The vehicle speed sensor and the yaw angular velocity sensor are both arranged on the vehicle, and are used to obtain the vehicle speed and yaw angular velocity of the vehicle respectively;

The first wheel and the second wheel are respectively arranged at two ends of the front axle, and the third wheel and the fourth wheel are respectively arranged at two ends of the rear axle;

The first to fourth wheel hub motors are respectively disposed on the first to fourth wheels, and are used to drive the first to fourth wheels;

The battery pack is arranged on the vehicle and is used for supplying power;

The vehicle electronic control unit is electrically connected to the steering wheel angle sensor, the vehicle speed sensor, the yaw angular velocity sensor, the motor controller, and the battery pack, respectively, and is used to calculate the torque of the four wheel hub motors according to the data measured by the steering wheel angle sensor, the vehicle speed sensor, and the yaw angular velocity sensor, and to generate corresponding current signals to be transmitted to the motor controller;

The motor controller is electrically connected to the four wheel hub motors and the battery respectively, and is used to control the operation of the four wheel hub motors according to the received current signals.

The present invention also discloses an adaptive neural network fault-tolerant control method based on the differential steering system, comprising the following steps:

Step 1), calculating the relationship between the ideal yaw rate and the steering wheel angle;

Step 2), establishing a state space model of the differential steering system;

Step 3), based on the state space model of the differential steering system, a state space model of the adaptive neural network fault-tolerant control system is established, and based on the state space model of the adaptive neural network fault-tolerant control system of the differential steering system, a state space model of the adaptive neural network fault-tolerant control system of the differential steering system is established in the event of a wheel hub motor failure;

Step 4), establishing a reference model and an inverse model of the adaptive neural network fault-tolerant control system;

Step 5), establishing a neural network compensator of the adaptive neural network fault-tolerant control system based on the reference model, the inverse model and the relationship between the ideal yaw angular velocity and the steering wheel angle of the adaptive neural network fault-tolerant control system;

Step 6), based on the neural network compensator of the adaptive neural network fault-tolerant control system, an adaptive neural network controller of the adaptive neural network fault-tolerant control system is established;

Step 7), based on the adaptive neural network controller, adaptively adjust the error between the reference model and the output of the adaptive neural network fault-tolerant control system when the hub motor fails.

As a further optimization scheme of the adaptive neural network fault-tolerant control method of a differential steering system of the present invention, the relationship between the ideal yaw angular velocity ω _r ^* and the steering wheel angle θ _sw in step 1) is:

a₀＝k_fk_r(a+b)²+(k_rb-k_fa)mu²；b₀＝k_fk_r(a+b)u；L为前后轴轴距；u为汽车速度；K_s为预设的横摆角速度调整参数；k_f、k_r分别为前后轮侧偏刚度；a为质心到前轴轴距；b为质心到后轴轴距；m为整车质量。Where:

a ₀ ＝k _f k _r (a+b) ² +(k _r bk _f a)mu ² ; b ₀ ＝k _f k _r (a+b)u; L is the wheelbase of the front and rear axles; u is the vehicle speed; K _s is the preset yaw rate adjustment parameter; k _f and k _r are the front and rear wheel cornering stiffnesses respectively; a is the wheelbase from the center of mass to the front axle; b is the wheelbase from the center of mass to the rear axle; m is the vehicle mass.

As a further optimization scheme of the adaptive neural network fault-tolerant control method of a differential steering system of the present invention, the state space model of the differential steering system in step 2) is:

In the formula,

_δf is the front wheel steering angle; β is the sideslip angle of the center of mass; _ωr is the yaw angular velocity; d is the semi-wheelbase; _Js is the equivalent moment of inertia of the steering wheel; G is the transmission ratio of the rack and pinion steering gear; I is the moment of inertia of the whole vehicle around the z-axis; _Bs is the equivalent damping of the steering wheel, R is the tire radius; _d2 is the tire drag moment; _d1 is the kingpin lateral offset moment; _Tsw is the torque applied by the driver on the steering wheel; _Tfl , _Tfr , _Trl , and _Trr are the output torques of the left front, right front, left rear, and right rear wheel hub motors, respectively.

As a further optimization scheme of the adaptive neural network fault-tolerant control method of a differential steering system of the present invention, the state space model of the adaptive neural network fault-tolerant control system of the differential steering system in step 3) is:

λ₁、λ₂、λ₃、λ₄分别为左前、右前、左后、右后轮毂电机发生故障的概率。Where, f(x(t)) = Ax(t); g(x(t)) = λB; h(x(t)) = Cx(t); t is time;

λ ₁ , λ ₂ , λ ₃ , and λ ₄ are the probabilities of failure of the left front, right front, left rear, and right rear wheel hub motors, respectively.

As a further optimization scheme of the adaptive neural network fault-tolerant control method of a differential steering system of the present invention, the state space model of the adaptive neural network fault-tolerant control system of the differential steering system described in step 3) when the wheel hub motor fails is:

Where σ(x(t),u(t),w(t)) is the disturbance input function under the hub motor fault.

As a further optimization scheme of the adaptive neural network fault-tolerant control method of a differential steering system of the present invention, the reference model of the adaptive neural network fault-tolerant control system described in step 4) is:

Where: _xm (t) is the state vector of the reference model; _um (t) is the input control vector of the reference model; _ym (t) is the output vector of the reference model; _Am =A; _Bm =λB; _Cm =C.

As a further optimization scheme of the adaptive neural network fault-tolerant control method of a differential steering system of the present invention, the inverse model of the adaptive neural network fault-tolerant control system described in step 4) is:

u(t)＝g ^-1 (t)[v(t)-f(x)]

Where: v(t) is the given tracking response.

As a further optimization scheme of the adaptive neural network fault-tolerant control method of a differential steering system of the present invention, the neural network compensator described in step 5) is:

Where: Δ is the inverse model error; _ys is the output of the s-th layer of the neural network; wis _is the weight from the ith neuron to the s-th layer of neurons; _gi (x) is the output value of the ith neuron; i is a natural number greater than or equal to 1 and less than or equal to n, n is the number of neurons, and s is the number of layers of the current neural network.

As a further optimization scheme of the adaptive neural network fault-tolerant control method of a differential steering system of the present invention, the adaptive neural network controller described in step 6) is:

Where: u _eer (t) is the compensation error of the inner loop system; K _p is the parameter matrix; u _NN is the output of the adaptive neural network controller.

Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

1. The electronic control unit collects steering wheel angle, yaw rate and vehicle speed signals in real time, calculates the difference between the ideal yaw rate and the actual yaw rate, recalculates the output torque of the wheel hub motor through the designed adaptive neural network controller, and sends current signals to the wheel hub motor by the ECU to complete the steering stability control under normal and failure conditions of the wheel hub motor;

2. The proposed control method is simple and reliable. At the same time, the neural network is used to effectively overcome the inverse model error and nonlinear factors caused by steering system failure, thereby realizing real-time following control of the differential steering model;

3. This method does not require prior knowledge of the location and size of the fault, nor does it require parameter identification of the system, but can ensure that the differential steering system accurately tracks the model output under fault conditions, thereby achieving ideal dynamic performance.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic structural diagram of a differential steering system in the present invention;

FIG2 is a schematic flow chart of the adaptive neural network fault-tolerant control method of the present invention.

In the figure, 1-steering wheel angle sensor, 2-steering wheel, 3-steering column, 4-left front wheel and hub motor, 5-gear rack steering gear, 6-right front wheel and hub motor, 7-front axle, 8-vehicle electronic control unit, 9-battery pack, 10-left rear wheel and hub motor, 11-right rear wheel and hub motor, 12-vehicle speed sensor, 13-yaw angular velocity sensor, 14-rear axle, 15-motor controller.

DETAILED DESCRIPTION

The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings:

As shown in FIG1 , the present invention develops a differential steering system, including a steering wheel angle sensor 1, a steering wheel 2, a steering column 3, a rack and pinion steering gear 5, first to fourth wheels, first to fourth wheel hub motors, a front axle 7, a vehicle electronic control unit 8, a battery pack 9, a vehicle speed sensor 12, a yaw rate sensor 13, a rear axle 14 and a motor control unit 15;

One end of the steering column 3 is fixedly connected to the steering wheel 2, and the other end is connected to the front axle 7 through a rack and pinion steering gear 5;

The steering wheel angle sensor 1 is arranged on the steering column 3 and is used to obtain the steering wheel angle;

The vehicle speed sensor 12 and the yaw angular velocity sensor 13 are both arranged on the vehicle, and are used to obtain the vehicle speed and yaw angular velocity of the vehicle respectively;

The first wheel and the second wheel are respectively arranged at two ends of the front axle 7, and the third wheel and the fourth wheel are respectively arranged at two ends of the rear axle 14;

The first to fourth wheel hub motors are respectively disposed on the first to fourth wheels, and are used to drive the first to fourth wheels;

The battery pack 9 is arranged on the vehicle for supplying power;

The vehicle electronic control unit 8 is electrically connected to the steering wheel angle sensor 1, the vehicle speed sensor 12, the yaw angular velocity sensor 13, the motor controller 15, and the battery pack 9, respectively, and is used to calculate the torque of the four wheel hub motors according to the data measured by the steering wheel angle sensor 1, the vehicle speed sensor 12, and the yaw angular velocity sensor 13, and generate corresponding current signals to be transmitted to the motor controller 15;

The motor controller 15 is electrically connected to the four wheel hub motors and the battery 9 respectively, and is used to control the operation of the four wheel hub motors according to the received current signal.

As shown in FIG2 , the present invention also discloses an adaptive neural network fault-tolerant control method based on the differential steering system, which is characterized by comprising the following steps:

Step 1), calculate the relationship between the ideal yaw rate ω _r ^* and the steering wheel angle θ _sw :

a₀＝k₁k₂(a+b)²+(k₂b-k₁a)mu²；b₀＝k₁k₂(a+b)u；L为前后轴轴距；u为汽车速度；K_s为预设的横摆角速度调整参数，其范围可根据驾驶员喜好选取，优先为0.12-0.37；k₁、k₂分别为前后轮侧偏刚度；a为质心到前轴轴距；b为质心到后轴轴距；m为整车质量。Where:

a ₀ ＝k ₁ k ₂ (a+b) ² +(k ₂ bk ₁ a)mu ² ；b ₀ ＝k ₁ k ₂ (a+b)u ；L is the wheelbase of the front and rear axles;u is the speed of the car;K _s is the preset yaw rate adjustment parameter, and its range can be selected according to the driver's preference, preferably 0.12-0.37;k ₁ and k ₂ are the front and rear wheel cornering stiffnesses respectively;a is the wheelbase from the center of mass to the front axle;b is the wheelbase from the center of mass to the rear axle;m is the vehicle mass.

Step 2), establish the state space model of the differential steering system:

The state space model of the differential steering system is:

In the formula,

_δf is the front wheel steering angle; β is the sideslip angle of the center of mass; _ωr is the yaw angular velocity; d is the semi-wheelbase; _Js is the equivalent moment of inertia of the steering wheel; G is the transmission ratio of the rack and pinion steering gear; I is the moment of inertia of the whole vehicle around the z-axis; _Bs is the equivalent damping of the steering wheel, R is the tire radius; _d2 is the tire drag moment; _d1 is the kingpin lateral offset moment; _Tsw is the torque applied by the driver on the steering wheel; _Tfl , _Tfr , _Trl , and _Trr are the output torques of the left front, right front, left rear, and right rear wheel hub motors, respectively.

Step 3), based on the state space model of the differential steering system, establish the state space model of its adaptive neural network fault-tolerant control system, and based on the state space model of the adaptive neural network fault-tolerant control system of the differential steering system, establish the state space model of the adaptive neural network fault-tolerant control system of the differential steering system in the event of a wheel hub motor failure.

First, the state space model of the adaptive neural network fault-tolerant control system of the differential steering system is established:

λ₁、λ₂、λ₃、λ₄分别为左前、右前、左后、右后轮毂电机发生故障的概率。Where, f(x(t)) = Ax(t); g(x(t)) = λB; h(x(t)) = Cx(t); t is time;

λ ₁ , λ ₂ , λ ₃ , and λ ₄ are the probabilities of failure of the left front, right front, left rear, and right rear wheel hub motors, respectively.

Based on the state space model of the adaptive neural network fault-tolerant control system of the differential steering system, when the differential system fails, the state space model of the differential steering system with adaptive neural network fault-tolerant function is:

Where, σ(x(t),u(t),w(t)) is the disturbance input function under fault conditions;

Step 4), establishing a reference model and an inverse model of the adaptive neural network fault-tolerant control system;

First, a reference model of the adaptive neural network fault-tolerant control system is established:

Where: _xm (t) is the state vector of the reference model; _um (t) is the input control vector of the reference model; _ym (t) is the output vector of the reference model; _Am =A; _Bm =λB; _Cm =C.

Secondly, the inverse model of the adaptive neural network fault-tolerant control system is established:

u(t)＝g ^-1 (t)[v(t)-f(x)]

Where: v(t) is the given tracking response.

Step 5), based on the reference model, inverse model and the relationship between the ideal yaw rate and the steering wheel angle of the above-mentioned adaptive neural network fault-tolerant control system, the neural network compensator of the adaptive neural network fault-tolerant control system is established, which can be expressed as:

Where: Δ is the inverse model error; _ys is the output of the s-th layer of the neural network; wis _is the weight from the ith neuron to the s-th layer of neurons; _gi (x) is the output value of the ith neuron; i is a natural number greater than or equal to 1 and less than or equal to n, n is the number of neurons, and s is the number of layers of the current neural network.

Step 6), based on the neural network compensator of the adaptive neural network fault-tolerant control system, an adaptive neural network controller of the adaptive neural network fault-tolerant control system is established as:

Where: u _eer (t) is the compensation error of the inner loop system; K _p is the parameter matrix; u _NN is the output of the adaptive neural network controller.

Step 7), based on the adaptive neural network regulator, the error between the reference model and the output of the adaptive neural network fault-tolerant control system under the hub motor fault is adaptively adjusted.

It can be seen from the structure diagram of the adaptive neural network controller of the present invention in FIG2 that the relationship between the inverse model error Δ and the input and system output of the inverse model is:

The following performance indicators are defined:

Where: y _js is the output of the j-th neuron; e _j is the error of the j-th neuron.

Firstly, the compensation error u _eer in various situations of offline training is used to make the output of the network close to u _eer , thereby completing the role of feedback compensation.

On the basis of offline learning, the data of differential steering system during operation is collected in real time, and the parameters are updated by online adaptive learning algorithm. In order to improve the stability of adaptive neural network fault-tolerant control system, the weights of neural network are adjusted, and the following online adaptive learning algorithm is adopted:

Where: W() is the weight; T, P are positive definite matrices; Q is the radial basis function.

During driving, the electronic control unit collects steering wheel angle, yaw rate and vehicle speed signals in real time, calculates the difference between the ideal yaw rate and the actual yaw rate, recalculates the output torque of the hub motor through the designed adaptive neural network controller, and sends a current signal to the hub motor by the motor control unit to complete the steering stability control under normal and failure conditions of the hub motor, thereby realizing a four-wheel steering system with adaptive neural network fault-tolerant control function and its control method.

It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as generally understood by those skilled in the art in the art to which the present invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with the meanings in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless defined as herein.

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

Claims (9)

1. An adaptive neural network fault-tolerant control method of a differential steering system comprises a steering wheel angle sensor (1), a steering wheel (2), a steering column (3), a rack-and-pinion steering gear (5), first to fourth wheels, first to fourth hub motors, a front axle (7), a whole vehicle electronic control unit (8), a storage battery pack (9), a vehicle speed sensor (12), a yaw rate sensor (13), a rear axle (14) and a motor controller (15);

one end of the steering column (3) is fixedly connected with the steering wheel (2), and the other end of the steering column is connected with the front shaft (7) through a rack-and-pinion steering gear (5);

the steering wheel angle sensor (1) is arranged on the steering column (3) and is used for acquiring the steering wheel angle;

the vehicle speed sensor (12) and the yaw rate sensor (13) are arranged on the vehicle and are respectively used for acquiring the vehicle speed and the yaw rate of the vehicle;

the first wheel and the second wheel are respectively arranged at two ends of the front axle (7), and the third wheel and the fourth wheel are respectively arranged at two ends of the rear axle (14);

the first to fourth hub motors are respectively and correspondingly arranged on the first to fourth wheels and are used for driving the first to fourth wheels;

the storage battery pack (9) is arranged on the automobile and is used for supplying power;

the whole vehicle electronic control unit (8) is respectively and electrically connected with the steering wheel angle sensor (1), the vehicle speed sensor (12), the yaw rate sensor (13), the motor controller (15) and the storage battery pack (9), and is used for calculating the moment of the four hub motors according to the data measured by the steering wheel angle sensor (1), the vehicle speed sensor (12) and the yaw rate sensor (13) and generating corresponding current signals to be transmitted to the motor controller (15);

the motor controller (15) is respectively and electrically connected with the four hub motors and the storage battery (9) and is used for controlling the four hub motors to work according to the received current signals;

the self-adaptive neural network fault-tolerant control method of the differential steering system is characterized by comprising the following steps of:

step 1), calculating the relation between the ideal yaw rate and the steering wheel angle;

step 2), establishing a state space model of the differential steering system;

step 3), a state space model of a self-adaptive neural network fault-tolerant control system of the differential steering system is established based on the state space model of the differential steering system, and a state space model of the self-adaptive neural network fault-tolerant control system of the differential steering system under the condition that a wheel hub motor fails is established based on the state space model of the self-adaptive neural network fault-tolerant control system of the differential steering system;

step 4), establishing a reference model and an inverse model of the adaptive neural network fault-tolerant control system;

step 5), building a neural network compensator of the adaptive neural network fault-tolerant control system based on a reference model, an inverse model and the relation between the ideal yaw rate and the steering wheel angle of the adaptive neural network fault-tolerant control system;

step 6), establishing an adaptive neural network controller of the adaptive neural network fault-tolerant control system based on a neural network compensator of the adaptive neural network fault-tolerant control system;

and 7) carrying out self-adaptive adjustment on the error between the reference model and the output of the self-adaptive neural network fault-tolerant control system under the fault of the hub motor based on the self-adaptive neural network controller.

2. The adaptive neural network fault-tolerant control method of a differential steering system according to claim 1, wherein the ideal yaw rate ω of step 1) is _r ^* And steering wheel angle theta _sw The relation is:

wherein:

a ₀ ＝k _f k _r (a+b) ² +(k _r b-k _f a)mu ² ；b ₀ ＝k _f k _r (a+b) u; l is the axial distance between the front axle and the rear axle; u is the speed of the car; k (K) _s Adjusting parameters for a preset yaw rate; k (k) _f 、k _r The lateral deflection rigidity of the front wheel and the rear wheel respectively; a is the axle distance from the mass center to the front axle; b is the axle distance from the mass center to the rear axle; m is the mass of the whole vehicle.

3. The adaptive neural network fault-tolerant control method of a differential steering system according to claim 2, wherein the state space model of the differential steering system in step 2) is:

in the method, in the process of the invention,

C＝[0 0 0 1]；

u(t)＝[T _fl T _fr T _rl T _rr ] ^T ；w(t)＝[T _sw ] ^T ；y(t)＝[ω _r ] ^T ；

θ _f is the front wheel corner; beta is the centroid slip angle; omega _r Is yaw rate; d is a half wheelbase; j (J) _s Equivalent moment of inertia for steering wheel; g is the gear ratio of the gear rack steering gear; i is the moment of inertia of the whole vehicle around the z axis; b (B) _s R is the tire radius, which is the equivalent damping of the steering wheel; d, d ₂ The tire drag torque; d, d ₁ A transverse offset moment for the kingpin; t (T) _sw Torque applied to the steering wheel for the driver; t (T) _fl 、T _fr 、T _rl 、T _rr The output torques of the front left, front right, rear left and rear right hub motors are respectively.

4. The method for adaptive neural network fault-tolerant control of a differential steering system according to claim 3, wherein the state space model of the adaptive neural network fault-tolerant control of a differential steering system in step 3) is:

where f (x (t))=ax (t); g (x (t))=λb; h (x (t))=cx (t); t is time;

λ ₁ 、λ ₂ 、λ ₃ 、λ ₄ the probability of failure of the front left, front right, rear left and rear right hub motors is respectively.

5. The adaptive neural network fault-tolerant control method of a differential steering system according to claim 4, wherein the state space model of the adaptive neural network fault-tolerant control system of the differential steering system in the case of a failure of the in-wheel motor in step 3) is:

where σ (x (t), u (t), w (t)) is a disturbance input function in the event of a hub motor failure.

6. The adaptive neural network fault-tolerant control method of a differential steering system according to claim 5, wherein the reference model of the adaptive neural network fault-tolerant control system in step 4) is:

wherein: x is x _m (t) is a state vector of the reference model; u (u) _m (t) input control vector for reference model, y _m (t) is the output vector of the reference model; a is that _m ＝A；B _m ＝λB；C _m ＝C。

7. The adaptive neural network fault-tolerant control method of a differential steering system according to claim 6, wherein the inverse model of the adaptive neural network fault-tolerant control system in step 4) is:

u(t)＝g ^-1 (t)[v(t)-f(x)]

wherein: v (t) is a given tracking response.

8. The adaptive neural network fault-tolerant control method of a differential steering system according to claim 7, wherein the neural network compensator in step 5) is:

wherein: delta is the inverse model error; y is _s Is the output of the s-th layer neural network; w (w) _is Weights for the ith neuron to the s-th layer neuron; g _i (x) Output value for the ith neuron; i is a natural number which is more than or equal to 1 and less than or equal to n, n is the number of neurons, and s is the number of layers of the current neural network.

9. The adaptive neural network fault-tolerant control method of a differential steering system according to claim 8, wherein the adaptive neural network controller in step 6) is:

wherein: u (u) _eer (t) is the compensation error of the inner loop system; k (K) _p Is a parameter matrix; u (u) _NN Is the output of the adaptive neural network controller.

Images