CN114454951A - Dual-motor steer-by-wire system and convolutional neural network fault-tolerant control method thereof - Google Patents

Dual-motor steer-by-wire system and convolutional neural network fault-tolerant control method thereof Download PDF

Info

Publication number
CN114454951A
CN114454951A CN202111646267.4A CN202111646267A CN114454951A CN 114454951 A CN114454951 A CN 114454951A CN 202111646267 A CN202111646267 A CN 202111646267A CN 114454951 A CN114454951 A CN 114454951A
Authority
CN
China
Prior art keywords
steering motor
steering
fault
motor
current
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202111646267.4A
Other languages
Chinese (zh)
Other versions
CN114454951B (en
Inventor
徐坤豪
赵万忠
张森皓
梁为何
刘津强
高犇
李志成
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Aeronautics and Astronautics
Original Assignee
Nanjing University of Aeronautics and Astronautics
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing University of Aeronautics and Astronautics filed Critical Nanjing University of Aeronautics and Astronautics
Priority to CN202111646267.4A priority Critical patent/CN114454951B/en
Publication of CN114454951A publication Critical patent/CN114454951A/en
Application granted granted Critical
Publication of CN114454951B publication Critical patent/CN114454951B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B62—LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
    • B62D—MOTOR VEHICLES; TRAILERS
    • B62D5/00—Power-assisted or power-driven steering
    • B62D5/04—Power-assisted or power-driven steering electrical, e.g. using an electric servo-motor connected to, or forming part of, the steering gear
    • B62D5/0457—Power-assisted or power-driven steering electrical, e.g. using an electric servo-motor connected to, or forming part of, the steering gear characterised by control features of the drive means as such
    • B62D5/046—Controlling the motor
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B62—LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
    • B62D—MOTOR VEHICLES; TRAILERS
    • B62D5/00—Power-assisted or power-driven steering
    • B62D5/04—Power-assisted or power-driven steering electrical, e.g. using an electric servo-motor connected to, or forming part of, the steering gear
    • B62D5/0457—Power-assisted or power-driven steering electrical, e.g. using an electric servo-motor connected to, or forming part of, the steering gear characterised by control features of the drive means as such
    • B62D5/046—Controlling the motor
    • B62D5/0463—Controlling the motor calculating assisting torque from the motor based on driver input
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B62—LAND VEHICLES FOR TRAVELLING OTHERWISE THAN ON RAILS
    • B62D—MOTOR VEHICLES; TRAILERS
    • B62D5/00—Power-assisted or power-driven steering
    • B62D5/04—Power-assisted or power-driven steering electrical, e.g. using an electric servo-motor connected to, or forming part of, the steering gear
    • B62D5/0457—Power-assisted or power-driven steering electrical, e.g. using an electric servo-motor connected to, or forming part of, the steering gear characterised by control features of the drive means as such
    • B62D5/0481—Power-assisted or power-driven steering electrical, e.g. using an electric servo-motor connected to, or forming part of, the steering gear characterised by control features of the drive means as such monitoring the steering system, e.g. failures
    • B62D5/0487—Power-assisted or power-driven steering electrical, e.g. using an electric servo-motor connected to, or forming part of, the steering gear characterised by control features of the drive means as such monitoring the steering system, e.g. failures detecting motor faults
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
    • H02H7/00—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions
    • H02H7/08—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions for dynamo-electric motors
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
    • H02H7/00—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions
    • H02H7/08—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions for dynamo-electric motors
    • H02H7/0822—Integrated protection, motor control centres
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/0003—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00—Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/0003—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control
    • H02P21/0014—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control using neural networks
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P23/00—Arrangements or methods for the control of AC motors characterised by a control method other than vector control
    • H02P23/0004—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P23/00—Arrangements or methods for the control of AC motors characterised by a control method other than vector control
    • H02P23/0004—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control
    • H02P23/0018—Control strategies in general, e.g. linear type, e.g. P, PI, PID, using robust control using neural networks
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P5/00—Arrangements specially adapted for regulating or controlling the speed or torque of two or more electric motors

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Chemical & Material Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Power Steering Mechanism (AREA)
  • Steering Control In Accordance With Driving Conditions (AREA)

Abstract

The invention discloses a double-motor steer-by-wire system and a convolutional neural network fault-tolerant control method thereof, wherein the system comprises the following steps: the steering wheel module, the steering execution module and the control module; the invention adopts a convolution neural network method to carry out real-time fault detection on the dual-motor steer-by-wire system, does not need to establish an accurate mathematical model, does not need an independent feature extraction stage, and improves the efficiency and the accuracy of the dual-motor fault detection; and the specific fault category of the double motors can be accurately diagnosed, and the safety and the stability of the double-motor steer-by-wire system are controlled through fault tolerance.

Description

Dual-motor steer-by-wire system and convolutional neural network fault-tolerant control method thereof
Technical Field
The invention belongs to the technical field of vehicle steer-by-wire systems, and particularly relates to a dual-motor steer-by-wire system and a convolutional neural network fault-tolerant control method thereof.
Background
As the steer-by-wire system gets rid of the connection of the traditional mechanical system and controls the operation of the whole system through electronic signals, for the steer-by-wire system with a single actuator, once the actuator has a problem, the whole system is crashed, and an active fault-tolerant introduction dual-motor structure is usually adopted. How to carry out real-time fault diagnosis on the dual-motor steer-by-wire system and quickly and accurately realize fault-tolerant control is the key for ensuring the stable operation of the dual-motor steer-by-wire system. Most of the existing fault diagnosis methods are directed to a single-motor system, and few researches are made on fault diagnosis of a dual-motor system, particularly a dual-motor steer-by-wire system.
The existing method for fault diagnosis and fault-tolerant control of the dual-motor steer-by-wire system is less disclosed, for example, the Chinese patent application No. CN201611018430.1 discloses a fault detection method applied to the dual-motor servo system, and fault diagnosis is carried out by modeling the system based on a Kalman filter observer; the Chinese patent application No. CN201110171716.4 discloses a redundant fault-tolerant control method applied to a dual-motor steer-by-wire system, which detects whether two motors have faults through a central controller so as to control the rotation angle of the normally working motor; the Chinese patent application No. CN201910136329.3 discloses a dual-motor dual-power-supply steer-by-wire system and a fault-tolerant control method thereof, which can carry out fast switching according to the fault conditions of a power supply and a motor through a plurality of power supply modes, and ensure that the other power supply drives the dual motors to carry out steering action under the condition of single power supply fault.
However, the fault-tolerant control method for the dual-motor steer-by-wire system mentioned in the above-mentioned prior patent is relatively simple, and only describes how to control the motor when a fault occurs, and does not consider how to perform fault diagnosis and real-time observation in combination with the system itself; and most observers are based on vehicle dynamics models or under the assumption that algorithm parameters are fixed, however, in practice most systems are highly complex, non-linear, which is difficult to observe by conventional observation methods. As known from literature, a key stage in the design of fault diagnosis tools is feature extraction. In convolutional neural networks, feature extraction and selection are part of the neural network, which makes the network more efficient in terms of both hardware and speed.
Disclosure of Invention
In view of the defects of the prior art, the invention aims to provide a dual-motor steer-by-wire system and a convolutional neural network fault-tolerant control method thereof, so as to solve the problems of inaccurate fault detection model, low accuracy and poor real-time performance in the prior art of the dual-motor steer-by-wire system.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
the invention relates to a dual-motor steer-by-wire system, comprising: the steering wheel module, the steering execution module and the control module;
a steering wheel module comprising: the device comprises a steering wheel, a steering column, a steering wheel corner sensor, a steering wheel torque sensor, a road sensing motor driver, a road sensing motor and a road sensing motor reducer;
the steering wheel is fixedly connected with one end of the steering column;
an output shaft of the road sensing motor is connected with the other end of the steering column through a road sensing motor reducer and used for transmitting road sensing to a steering wheel through the steering column;
the road sensing motor driver is connected with the road sensing motor and used for driving the rotation state of the road sensing motor;
the steering wheel corner sensor and the steering wheel torque sensor are fixedly connected with the steering column, respectively collect corner and torque signals of the steering wheel and send the collected signals to the control module;
a steering execution module comprising: the device comprises a first steering motor, a first steering motor reducer, a first steering motor driver, a first pinion, a second steering motor reducer, a second steering motor driver, a second pinion, a rack, a steering tie rod, a front wheel, a first current Hall sensor, a second current Hall sensor, a rack displacement sensor, a vehicle speed sensor and a yaw rate sensor;
the first steering motor is connected with a rotating shaft of the first pinion through a first steering motor reducer, and the second steering motor is connected with a rotating shaft of the second pinion through a second steering motor reducer;
the first pinion and the second pinion are meshed with the rack; the rack is connected with the steering tie rod; two ends of the steering tie rod are respectively connected with two front wheels of the vehicle;
the vehicle speed sensor is arranged in the front wheel and used for acquiring the vehicle speed of the vehicle and sending the vehicle speed to the control module;
the yaw rate sensor is arranged on the vehicle body and used for obtaining an actual yaw rate signal of the vehicle and sending the actual yaw rate signal to the control module;
the first current Hall sensor and the second current Hall sensor are respectively arranged on the first steering motor and the second steering motor and are used for acquiring a first steering motor current signal and a second steering motor current signal and sending the first steering motor current signal and the second steering motor current signal to the control module;
the rack displacement sensor is arranged on the rack and the steering tie rod and used for detecting a displacement signal of the rack and sending the displacement signal to the control module;
the first steering motor driver is electrically connected with the first steering motor and the control module respectively and used for controlling the rotation state of the first steering motor; the second steering motor driver is electrically connected with the second steering motor and the control module respectively and used for controlling the rotation state of the second steering motor;
the control module is respectively and electrically connected with a steering wheel angle sensor, a steering wheel torque sensor, a vehicle speed sensor, a yaw rate sensor, a first current Hall sensor, a second current Hall sensor, a rack displacement sensor, a road sensing motor driver, a first steering motor driver and a second steering motor driver.
Further, the control module includes: the fault diagnosis system comprises an information acquisition module, a fault diagnosis module and a fault-tolerant control module;
the information acquisition module is used for filtering and denoising the acquired steering wheel corner signal, steering wheel torque signal, vehicle speed signal, yaw rate signal, first steering motor current signal, second steering motor current signal and rack displacement signal, and sending the processed signals to the fault diagnosis module;
the fault diagnosis module carries out fault diagnosis on the first steering motor and the second steering motor in real time through an effective convolutional neural network model according to the signal sent by the information acquisition module, and transmits a generated fault vector label of the steering motor to the fault-tolerant control module;
and the fault-tolerant control module judges the fault category and the fault condition of the steering motor according to the fault vector label of the steering motor, and respectively performs fault-tolerant control on the first steering motor and the second steering motor according to different fault conditions.
Further, the models of the first steering motor and the second steering motor are the same.
Further, the first pinion and the second pinion are the same in model.
The invention discloses a convolutional neural network fault-tolerant control method of a double-motor steer-by-wire system, which is based on the system and comprises the following steps:
1) collecting current signals of a first steering motor and a second steering motor and rack displacement signals of the dual-motor steer-by-wire system in different working states to form a total sample, randomly dividing the total sample into a training sample and a testing sample, performing category marking, and forming a fault vector label corresponding to normal and fault conditions of resistance and moment coefficient of the steering motor respectively;
2) establishing a convolutional neural network model, training the convolutional neural network model by using a marked training sample, and inputting a marked test sample into the trained convolutional neural network model for verification to obtain an effective convolutional neural network model;
3) performing real-time fault diagnosis on the first steering motor and the second steering motor through an effective convolutional neural network model to obtain current fault vector labels of the first steering motor and the second steering motor;
4) and carrying out fault-tolerant control on the dual-motor steer-by-wire system according to the current fault vector labels of the first steering motor and the second steering motor obtained in the step 3).
Further, the fault vector label in step 1) specifically includes: the resistance of the first steering motor is normal, and the resistance of the second steering motor is failed; the first steering motor resistor fails, and the second steering motor resistor is normal; a first steering motor resistance fault and a second steering motor resistance fault; the torque coefficient of the first steering motor is in fault, and the torque coefficient of the second steering motor is normal; the torque coefficient of the first steering motor is normal, and the torque coefficient of the second steering motor is failed; the first steering motor moment coefficient fault and the second steering motor moment coefficient fault; the first steering motor is all normal, and the second steering motor is all normal.
Further, the resistance fault conditions of the first steering motor and the second steering motor are marked according to the first steering motor current signal and the second steering motor current signal in the step 1); and marking the moment coefficient fault conditions of the first steering motor and the second steering motor according to the current signal of the first steering motor, the current signal of the second steering motor and the rack displacement signal.
Further, the establishing of the convolutional neural network model in the step 2) specifically includes:
21) the convolutional layer convolves an input signal array with a group of filters with different sizes, the training speed is increased through batch normalization, and a target output characteristic diagram is generated by using a ReLU activation function in the same layer; the feature map for convolutional layer extraction is represented as:
Figure BDA0003445293490000031
in the formula (I), the compound is shown in the specification,
Figure BDA0003445293490000041
for the jth output signature of the n convolutional layers,
Figure BDA0003445293490000042
is the input characteristic diagram of the (n-1) th convolutional layer,
Figure BDA0003445293490000043
for the convolution kernel connecting the ith input feature map and the jth output feature map in the nth convolution layer,
Figure BDA0003445293490000044
deviation of nth layer, representing two-dimensional convolution operation, MjIs an input feature map set, and f is a ReLU activation function;
22) the pool layer minimizes its dimensions by modifying the convolutional layer extracted feature map to a single output:
Figure BDA0003445293490000045
in the formula (I), the compound is shown in the specification,
Figure BDA0003445293490000046
is the output of the nth pool layer; lB、WBThe length and width of the window;
23) the fully-connected layer obtains output feature mapping generated by the convolutional layer and the pool layer, and classifies input data into labels through the output feature mapping:
Figure BDA0003445293490000047
in the formula, WfAnd BfThe weight and deviation of the fully connected layer respectively;
24) and (3) carrying out fault classification on the softmax layer, and outputting a fault vector O as:
Figure BDA0003445293490000048
where O is an output fault vector, Y is a feature type, X is an input signal at the present time, W is a weight of each of 7 types, and b is a deviation of each of 7 types.
Further, the process of training the convolutional neural network model in step 2) specifically includes: and adjusting various weights and deviation parameters of the convolutional neural network model according to the cross entropy errors of the network estimation value and the label by adopting a back propagation method to obtain an effective convolutional neural network model.
Further, the fault-tolerant control adopted in the step 4) specifically includes:
41) if the current fault vector label is that the resistance of the first steering motor is normal and the resistance of the second steering motor is in fault, the current of the second steering motor is cut off, and the first steering motor passes through H2/HinfThe algorithm carries out corner control and tracks the ideal front wheel corner; reminding a driver of replacing the second steering motor in time;
42) if the current fault vector label is that the first steering motor resistor is in fault and the second steering motor resistor is normal, the current of the first steering motor is cut off, and the second steering motor passes through H2/HinfThe algorithm carries out corner control and tracks the ideal front wheel corner; reminding a driver of replacing the first steering motor in time;
43) if the current fault vector label is a first steering motor moment coefficient fault and a second steering motor moment coefficient, the first steering motor moment coefficient fault occurs, and the second steering motor moment coefficient is normal; the first steering motor performs corner control through a mu control algorithm, and the second steering motor performs torque control through a PID control algorithm; reminding a driver of replacing the first steering motor in time;
44) if the current fault vector label is that the torque coefficient of the first steering motor is normal and the torque coefficient of the second steering motor is normal, the second steering motor performs corner control through a mu control algorithm, and the first steering motor performs torque control through a PID control algorithm; reminding a driver of replacing the second steering motor in time;
45) if the current fault vector labels are a first steering motor resistance fault and a second steering motor resistance fault, cutting off the current of the first steering motor and the second steering motor, stopping controlling the two steering motors, reminding a driver to stop parking along the side and timely replacing the first steering motor and the second steering motor;
46) if the current fault vector labels are a first steering motor moment coefficient fault and a second steering motor moment coefficient fault, the control voltages of the first steering motor and the second steering motor are increased, and the two steering motors are controlled through a mu control algorithm; reminding a driver of replacing the first steering motor and the second steering motor in time;
47) if the current fault vector labels are that the first steering motor is all normal and the second steering motor is all normal, the first steering motor and the second steering motor both pass through H2/HinfThe algorithm performs corner control and tracks the ideal front wheel corner.
Further, said H2/HinfThe control algorithm is as follows:
the state variable of the control system is
Figure BDA0003445293490000051
The input is u ═ Δ T]The measurement output is y1＝[γ]，y2＝[ΔT]The state space of the yaw-rate compensation control based on the active front-wheel steering is implemented as follows:
Figure BDA0003445293490000052
in the formula:
Figure BDA0003445293490000061
Figure BDA0003445293490000062
C11＝[0 0 0 1]，D11＝[0]；C12＝[0 0 0 0]；D12＝[1]
in the formula,. DELTA.theta.sIs the compensated pinion angle, Δ T is the compensation torque of the normal motor, m is the vehicle mass, k1,k2Respectively the cornering stiffness of the front and rear tires, a, b respectively the distance from the center of mass to the front and rear axles, u is the vehicle speed, gamma is the vehicle yaw rate, beta is the vehicle center of mass cornering angle, IzFor the moment of inertia, delta, of the finished vehicle about the z-axisfAt a corner of the front wheel, JRIs rack equivalent moment of inertia, BRIs rack equivalent damping coefficient, G1For reduction ratio of two steering motor reducers, G2Is the reduction ratio of the rack-and-pinion mechanism, eta is the efficiency coefficient of the two steering motor reducers, tp，tmRespectively, tire drag and kingpin offset, JmIs the rotational inertia of the motor, BmIs the motor damping coefficient.
Further, the μ control algorithm is as follows:
taking state variables of control systems
Figure BDA0003445293490000063
Input u ═ Δ T to the system]The disturbance input of the system is w ═ dr Fyw]T，y＝[γ]The system output, the state space of the drive-by-wire front wheel steering yaw velocity control is realized as follows:
Figure BDA0003445293490000064
in the formula:
Figure BDA0003445293490000071
Figure BDA0003445293490000072
C＝[0 0 0 0 0 1]；D1＝[0 0]；D2＝[0]
in the formula, thetas1Is the current total pinion angle; thetas2Is the total compensation angle of the pinion under the action of the mu controller; Δ T is the total compensation torque of the pinion under the action of the fault-tolerant controller; drIs road surface disturbance; fywIs a side wind disturbance.
The invention has the beneficial effects that:
1. the invention adopts a convolution neural network method to carry out real-time fault detection on the dual-motor steer-by-wire system, does not need to establish an accurate mathematical model, does not need an independent feature extraction stage, and improves the efficiency and the accuracy of the dual-motor fault detection; the specific fault category (resistance fault or moment coefficient fault) of the double motors can be accurately diagnosed, and the safety and the stability of the double-motor steer-by-wire system are controlled through fault tolerance;
2. for the fault-tolerant control problem of the dual-motor steer-by-wire system, the invention considers the stability of the vehicle while carrying out fault-tolerant control according to the fault condition diagnosed by the dual motors in real time, thereby ensuring the safety of the system and the driver.
Drawings
FIG. 1 is a schematic diagram of the system of the present invention;
FIG. 2 is a flow chart of the method of the present invention;
fig. 3 is a diagram of verifying the accuracy of diagnosing a convolutional neural network failure according to an embodiment of the present invention.
Detailed Description
In order to facilitate understanding of those skilled in the art, the present invention will be further described with reference to the following examples and drawings, which are not intended to limit the present invention.
Referring to fig. 1, a dual-motor steer-by-wire system of the present invention includes: the steering wheel module, the steering execution module and the control module 5;
a steering wheel module comprising: the device comprises a steering wheel 1, a steering column 2, a steering wheel angle sensor 3, a steering wheel torque sensor 19, a road sensing motor driver 17, a road sensing motor 18 and a road sensing motor reducer 4;
the steering wheel 1 is fixedly connected with one end of a steering column 2;
an output shaft of the road feel motor 18 is connected with the other end of the steering column 2 through a road feel motor reducer 4 and is used for transmitting road feel to the steering wheel 1 through the steering column 2;
the road sensing motor driver 17 is connected with the road sensing motor 18 and is used for driving the rotation state of the road sensing motor 18;
the steering wheel corner sensor 3 and the steering wheel torque sensor 19 are fixedly connected with the steering column 2, respectively collect corner and torque signals of the steering wheel 1 and send the collected signals to the control module;
a steering execution module comprising: a first steering motor 8, a first steering motor reducer 9, a first steering motor driver 6, a first pinion 10, a second steering motor 15, a second steering motor reducer 14, a second steering motor driver 16, a second pinion 11, a rack 12, a tie rod 20, a front wheel 7, a first current hall sensor, a second current hall sensor, a rack displacement sensor 13, a vehicle speed sensor, and a yaw rate sensor; the first steering motor and the second steering motor are the same in model; the first pinion and the second pinion are the same in model.
The first steering motor 8 is connected with a rotating shaft of the first pinion 10 through a first steering motor reducer 9, and the second steering motor 15 is connected with a rotating shaft of the second pinion 11 through a second steering motor reducer 14;
the first pinion 10 and the second pinion 11 are both meshed with the rack 12; the rack 12 is connected with a tie rod 20; both ends of the tie rod 20 are connected to the two front wheels 7 of the vehicle, respectively;
the vehicle speed sensor is arranged in the front wheel and used for acquiring the vehicle speed of the vehicle and sending the vehicle speed to the control module;
the yaw rate sensor is arranged on the vehicle body and used for obtaining an actual yaw rate signal of the vehicle and sending the actual yaw rate signal to the control module;
the first current Hall sensor and the second current Hall sensor are respectively arranged on the first steering motor 8 and the second steering motor 15 and are used for acquiring a first steering motor current signal and a second steering motor current signal and sending the first steering motor current signal and the second steering motor current signal to the control module;
the rack displacement sensor 13 is mounted on the rack 12 and the tie rod 20, and is used for detecting a displacement signal of the rack 12 and sending the displacement signal to the control module;
the first steering motor driver 6 is electrically connected with the first steering motor 8 and the control module 5 respectively and is used for controlling the rotation state of the first steering motor 8; the second steering motor driver 16 is electrically connected with the second steering motor 15 and the control module 5 respectively and is used for controlling the rotation state of the second steering motor 15;
the control module 5 is respectively electrically connected with the steering wheel angle sensor 3, the steering wheel torque sensor 19, the vehicle speed sensor, the yaw rate sensor, the first current Hall sensor, the second current Hall sensor, the rack displacement sensor 13, the road sensing motor driver 17, the first steering motor driver and the second steering motor driver;
wherein the control module comprises: the fault diagnosis system comprises an information acquisition module, a fault diagnosis module and a fault-tolerant control module;
the information acquisition module is used for filtering and denoising the acquired steering wheel angle signal, the steering wheel torque signal, the vehicle speed signal, the yaw velocity signal, the first steering motor current signal, the second steering motor current signal and the rack displacement signal and sending the processed signals to the fault diagnosis module;
the fault diagnosis module carries out fault diagnosis on the first steering motor and the second steering motor in real time through an effective convolutional neural network model according to the signal sent by the information acquisition module, and transmits a generated fault vector label of the steering motor to the fault-tolerant control module;
and the fault-tolerant control module judges the fault category and the fault condition of the steering motor according to the fault vector label of the steering motor, and respectively performs fault-tolerant control on the first steering motor and the second steering motor according to different fault conditions.
Referring to fig. 2, the convolutional neural network fault-tolerant control method of the dual-motor steer-by-wire system of the present invention, based on the above system, includes the following steps:
1) collecting current signals of a first steering motor and a second steering motor and rack displacement signals of the dual-motor steer-by-wire system in different working states to form a total sample, randomly dividing the total sample into a training sample and a testing sample, performing category marking, and forming a fault vector label corresponding to normal and fault conditions of resistance and moment coefficient of the steering motor respectively;
wherein, the fault vector label in step 1) specifically includes: the resistance of the first steering motor is normal, and the resistance of the second steering motor is failed; the first steering motor resistor fails, and the second steering motor resistor is normal; a first steering motor resistance fault and a second steering motor resistance fault; the torque coefficient of the first steering motor is in fault, and the torque coefficient of the second steering motor is normal; the torque coefficient of the first steering motor is normal, and the torque coefficient of the second steering motor is failed; the first steering motor moment coefficient fault and the second steering motor moment coefficient fault; the first steering motor is all normal, and the second steering motor is all normal.
Marking resistance fault conditions of the first steering motor and the second steering motor according to the current signal of the first steering motor and the current signal of the second steering motor in the step 1); and marking the moment coefficient fault conditions of the first steering motor and the second steering motor according to the current signal of the first steering motor, the current signal of the second steering motor and the rack displacement signal.
2) Establishing a convolutional neural network model, training the convolutional neural network model by using a marked training sample, and inputting a marked test sample into the trained convolutional neural network model for verification to obtain an effective convolutional neural network model;
specifically, the establishing of the convolutional neural network model in the step 2) specifically includes:
21) the convolutional layer convolves an input signal array with a group of filters with different sizes, the training speed is increased through batch normalization, and a target output characteristic diagram is generated by using a ReLU activation function in the same layer; the feature map for convolutional layer extraction is represented as:
Figure BDA0003445293490000091
in the formula (I), the compound is shown in the specification,
Figure BDA0003445293490000092
for the jth output profile of the n convolutional layers,
Figure BDA0003445293490000093
is the input characteristic diagram of the (n-1) th convolutional layer,
Figure BDA0003445293490000094
for the convolution kernel connecting the ith input feature map and the jth output feature map in the nth convolution layer,
Figure BDA0003445293490000095
deviation of nth layer, representing two-dimensional convolution operation, MjIs an input feature map set, and f is a ReLU activation function;
22) the pool layer minimizes its dimensions by modifying the convolutional layer extracted feature map to a single output:
Figure BDA0003445293490000101
in the formula (I), the compound is shown in the specification,
Figure BDA0003445293490000102
is the output of the nth pool layer; lB、WBThe length and width of the window;
23) the fully-connected layer obtains output feature mapping generated by the convolutional layer and the pool layer, and classifies input data into labels through the output feature mapping:
Figure BDA0003445293490000103
in the formula, WfAnd BfThe weight and deviation of the fully connected layer respectively;
24) and (3) carrying out fault classification on the softmax layer, and outputting a fault vector O as:
Figure BDA0003445293490000104
where O is an output fault vector, Y is a feature type, X is an input signal at the present time, W is a weight of each of 7 types, and b is a deviation of each of 7 types.
The process of training the convolutional neural network model specifically comprises the following steps: and adjusting various weights and deviation parameters of the convolutional neural network model according to the cross entropy errors of the network estimation value and the label by adopting a back propagation method to obtain an effective convolutional neural network model.
Referring to fig. 3, the effective convolutional neural network model obtained through training is verified through the collected test samples, the accuracy rate reaches 97.75%, and the test samples are proved to be effective and can be used for fault diagnosis of the dual-motor steer-by-wire system.
3) Performing real-time fault diagnosis on the first steering motor and the second steering motor through an effective convolutional neural network model to obtain current fault vector labels of the first steering motor and the second steering motor;
4) and carrying out fault-tolerant control on the dual-motor steer-by-wire system according to the current fault vector labels of the first steering motor and the second steering motor obtained in the step 3).
Specifically, the fault-tolerant control adopted in step 4) specifically includes:
41) if the current fault vector label is that the resistance of the first steering motor is normal and the resistance of the second steering motor is in fault, the current of the second steering motor is cut off, and the first steering motor passes through H2/HinfThe algorithm carries out corner control and tracks the ideal front wheel corner; and liftWaking the driver to replace the second steering motor in time;
42) if the current fault vector label is that the first steering motor resistor is in fault and the second steering motor resistor is normal, the current of the first steering motor is cut off, and the second steering motor passes through H2/HinfThe algorithm carries out corner control and tracks the ideal front wheel corner; reminding a driver of replacing the first steering motor in time;
43) if the current fault vector label is a first steering motor moment coefficient fault and a second steering motor moment coefficient, the first steering motor moment coefficient fault occurs, and the second steering motor moment coefficient is normal; the first steering motor performs corner control through a mu control algorithm, and the second steering motor performs torque control through a PID control algorithm; reminding a driver of replacing the first steering motor in time;
44) if the current fault vector label is that the torque coefficient of the first steering motor is normal and the torque coefficient of the second steering motor is normal, the second steering motor performs corner control through a mu control algorithm, and the first steering motor performs torque control through a PID control algorithm; reminding a driver of replacing the second steering motor in time;
45) if the current fault vector labels are a first steering motor resistance fault and a second steering motor resistance fault, cutting off the current of the first steering motor and the second steering motor, stopping controlling the two steering motors, reminding a driver to stop parking along the side and timely replacing the first steering motor and the second steering motor;
46) if the current fault vector labels are a first steering motor moment coefficient fault and a second steering motor moment coefficient fault, the control voltages of the first steering motor and the second steering motor are increased, and the two steering motors are controlled through a mu control algorithm; reminding a driver of replacing the first steering motor and the second steering motor in time;
47) if the current fault vector labels are that the first steering motor is all normal and the second steering motor is all normal, the first steering motor and the second steering motor both pass through H2/HinfThe algorithm performs corner control and tracks the ideal front wheel corner.
Wherein, the H2/HinfThe control algorithm is as follows:
the state variable of the control system is
Figure BDA0003445293490000111
The input is u ═ Δ T]The measurement output is y1＝[γ]，y2＝[ΔT]The state space of the yaw-rate compensation control based on the active front-wheel steering is implemented as follows:
Figure BDA0003445293490000112
in the formula:
Figure BDA0003445293490000121
Figure BDA0003445293490000122
C11＝[0 0 0 1]，D11＝[0]；C12＝[0 0 0 0]；D12＝[1]
in the formula,. DELTA.theta.sIs the compensated pinion angle, Δ T is the compensation torque of the normal motor, m is the vehicle mass, k1,k2Respectively the cornering stiffness of the front and rear tires, a, b respectively the distance from the center of mass to the front and rear axles, u is the vehicle speed, gamma is the vehicle yaw rate, beta is the vehicle center of mass cornering angle, IzFor the moment of inertia, delta, of the finished vehicle about the z-axisfAt a corner of the front wheel, JRIs rack equivalent moment of inertia, BRIs rack equivalent damping coefficient, G1For reduction ratio of two steering motor reducers, G2Is the reduction ratio of the rack-and-pinion mechanism, eta is the efficiency coefficient of the two steering motor reducers, tp，tmRespectively, tire drag and kingpin offset, JmIs the rotational inertia of the motor, BmIs the motor damping coefficient.
Wherein the mu control algorithm is as follows:
fetch and controlState variables of system
Figure BDA0003445293490000123
Input u ═ Δ T to the system]The disturbance input of the system is w ═ dr Fyw]T，y＝[γ]The system output, the state space of the drive-by-wire front wheel steering yaw velocity control is realized as follows:
Figure BDA0003445293490000124
in the formula:
Figure BDA0003445293490000131
Figure BDA0003445293490000132
C＝[0 0 0 0 0 1]；D1＝[0 0]；D2＝[0]
in the formula, thetas1Is the current total pinion angle; thetas2Is the total compensation angle of the pinion under the action of the mu controller; Δ T is the total compensation torque of the pinion under the action of the fault-tolerant controller; drIs road surface interference; fywIs a side wind disturbance.
While the invention has been described in terms of its preferred embodiments, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.

Claims (10)

1. A dual motor steer-by-wire system, comprising: the steering wheel module, the steering execution module and the control module;
a steering wheel module comprising: the device comprises a steering wheel, a steering column, a steering wheel corner sensor, a steering wheel torque sensor, a road sensing motor driver, a road sensing motor and a road sensing motor reducer;
the steering wheel is fixedly connected with one end of the steering column;
an output shaft of the road sensing motor is connected with the other end of the steering column through a road sensing motor reducer and used for transmitting road sensing to a steering wheel through the steering column;
the road sensing motor driver is connected with the road sensing motor and used for driving the rotation state of the road sensing motor;
the steering wheel corner sensor and the steering wheel torque sensor are fixedly connected with the steering column, respectively collect corner and torque signals of the steering wheel and send the collected signals to the control module;
a steering execution module comprising: the device comprises a first steering motor, a first steering motor reducer, a first steering motor driver, a first pinion, a second steering motor reducer, a second steering motor driver, a second pinion, a rack, a steering tie rod, a front wheel, a first current Hall sensor, a second current Hall sensor, a rack displacement sensor, a vehicle speed sensor and a yaw rate sensor;
the first steering motor is connected with a rotating shaft of the first pinion through a first steering motor reducer, and the second steering motor is connected with a rotating shaft of the second pinion through a second steering motor reducer;
the first pinion and the second pinion are meshed with the rack; the rack is connected with the steering tie rod; two ends of the steering tie rod are respectively connected with two front wheels of the vehicle;
the vehicle speed sensor is arranged in the front wheel and used for acquiring the vehicle speed of the vehicle and sending the vehicle speed to the control module;
the yaw rate sensor is arranged on the vehicle body and used for obtaining an actual yaw rate signal of the vehicle and sending the actual yaw rate signal to the control module;
the first current Hall sensor and the second current Hall sensor are respectively arranged on the first steering motor and the second steering motor and are used for acquiring a first steering motor current signal and a second steering motor current signal and sending the first steering motor current signal and the second steering motor current signal to the control module;
the rack displacement sensor is arranged on the rack and the steering tie rod and used for detecting a displacement signal of the rack and sending the displacement signal to the control module;
the first steering motor driver is electrically connected with the first steering motor and the control module respectively and used for controlling the rotation state of the first steering motor; the second steering motor driver is electrically connected with the second steering motor and the control module respectively and used for controlling the rotation state of the second steering motor;
the control module is respectively and electrically connected with a steering wheel angle sensor, a steering wheel torque sensor, a vehicle speed sensor, a yaw rate sensor, a first current Hall sensor, a second current Hall sensor, a rack displacement sensor, a road sensing motor driver, a first steering motor driver and a second steering motor driver.
2. The dual-motor steer-by-wire system of claim 1, wherein the control module comprises: the fault diagnosis system comprises an information acquisition module, a fault diagnosis module and a fault-tolerant control module;
the information acquisition module is used for filtering and denoising the acquired steering wheel corner signal, steering wheel torque signal, vehicle speed signal, yaw rate signal, first steering motor current signal, second steering motor current signal and rack displacement signal, and sending the processed signals to the fault diagnosis module;
the fault diagnosis module carries out fault diagnosis on the first steering motor and the second steering motor in real time through an effective convolutional neural network model according to the signal sent by the information acquisition module, and transmits a generated fault vector label of the steering motor to the fault-tolerant control module;
the fault-tolerant control module judges the fault type and the fault condition of the steering motor according to the fault vector label of the steering motor, and respectively carries out fault-tolerant control on the first steering motor and the second steering motor according to different fault conditions.
3. The dual-motor steer-by-wire system of claim 1, wherein the first and second steer motors are the same size.
4. The dual-motor steer-by-wire system of claim 1, wherein the first and second pinions are of the same size.
5. A convolutional neural network fault-tolerant control method of a dual-motor steer-by-wire system, which is based on the system of any one of claims 1 to 4, and is characterized by comprising the following steps:
1) collecting current signals of a first steering motor and a second steering motor and rack displacement signals of the dual-motor steer-by-wire system in different working states to form a total sample, randomly dividing the total sample into a training sample and a testing sample, performing category marking, and forming a fault vector label corresponding to normal and fault conditions of resistance and moment coefficient of the steering motor respectively;
2) establishing a convolutional neural network model, training the convolutional neural network model by using a marked training sample, and inputting a marked test sample into the trained convolutional neural network model for verification to obtain an effective convolutional neural network model;
3) performing real-time fault diagnosis on the first steering motor and the second steering motor through an effective convolutional neural network model to obtain current fault vector labels of the first steering motor and the second steering motor;
4) and carrying out fault-tolerant control on the dual-motor steer-by-wire system according to the current fault vector labels of the first steering motor and the second steering motor obtained in the step 3).
6. The convolutional neural network fault-tolerant control method of a dual-motor steer-by-wire system according to claim 5, wherein the fault vector tag in the step 1) specifically comprises: the resistance of the first steering motor is normal, and the resistance of the second steering motor is failed; the first steering motor resistor fails, and the second steering motor resistor is normal; a first steering motor resistance fault and a second steering motor resistance fault; the torque coefficient of the first steering motor is in fault, and the torque coefficient of the second steering motor is normal; the torque coefficient of the first steering motor is normal, and the torque coefficient of the second steering motor is failed; the first steering motor moment coefficient fault and the second steering motor moment coefficient fault; the first steering motor is all normal, and the second steering motor is all normal.
7. The convolutional neural network fault-tolerant control method of a two-motor steer-by-wire system of claim 6, wherein the resistive fault condition of the first steering motor and the second steering motor is marked in step 1) according to the first steering motor current signal and the second steering motor current signal; and marking the moment coefficient fault conditions of the first steering motor and the second steering motor according to the current signal of the first steering motor, the current signal of the second steering motor and the rack displacement signal.
8. The convolutional neural network fault-tolerant control method of the dual-motor steer-by-wire system according to claim 5, wherein the establishing of the convolutional neural network model in the step 2) specifically comprises:
21) the convolutional layer convolves an input signal array with a group of filters with different sizes, the training speed is increased through batch normalization, and a target output characteristic diagram is generated by using a ReLU activation function in the same layer; the feature map for convolutional layer extraction is represented as:
Figure FDA0003445293480000031
in the formula (I), the compound is shown in the specification,
Figure FDA0003445293480000032
for the jth output signature of the n convolutional layers,
Figure FDA0003445293480000033
is the input characteristic diagram of the (n-1) th convolutional layer,
Figure FDA0003445293480000034
for the convolution kernel connecting the ith input feature map and the jth output feature map in the nth convolution layer,
Figure FDA0003445293480000035
deviation of nth layer, representing two-dimensional convolution operation, MjIs an input feature map set, and f is a ReLU activation function;
22) the pool layer minimizes its dimensions by modifying the convolutional layer extracted feature map to a single output:
Figure FDA0003445293480000036
in the formula (I), the compound is shown in the specification,
Figure FDA0003445293480000037
is the output of the nth pool layer; lB、WBThe length and width of the window;
23) the fully-connected layer obtains output feature mapping generated by the convolutional layer and the pool layer, and classifies input data into labels through the output feature mapping:
Figure FDA0003445293480000038
in the formula, WfAnd BfThe weight and deviation of the fully connected layer respectively;
24) and (3) carrying out fault classification on the softmax layer, and outputting a fault vector O as:
Figure FDA0003445293480000041
where O is an output fault vector, Y is a feature type, X is an input signal at the present time, W is a weight of each of 7 types, and b is a deviation of each of 7 types.
9. The convolutional neural network fault-tolerant control method of the dual-motor steer-by-wire system according to claim 5, wherein the process of training the convolutional neural network model in the step 2) specifically comprises the following steps: and adjusting various weights and deviation parameters of the convolutional neural network model according to the cross entropy errors of the network estimation value and the label by adopting a back propagation method to obtain an effective convolutional neural network model.
10. The convolutional neural network fault-tolerant control method of a two-motor steer-by-wire system according to claim 5, wherein the fault-tolerant control adopted in the step 4) specifically comprises:
41) if the current fault vector label is that the resistance of the first steering motor is normal and the resistance of the second steering motor is in fault, the current of the second steering motor is cut off, and the first steering motor passes through H2/HinfThe algorithm carries out corner control and tracks the ideal front wheel corner; reminding a driver of replacing the second steering motor in time;
42) if the current fault vector label is that the first steering motor resistor is in fault and the second steering motor resistor is normal, the current of the first steering motor is cut off, and the second steering motor passes through H2/HinfThe algorithm carries out corner control and tracks the ideal front wheel corner; reminding a driver of replacing the first steering motor in time;
43) if the current fault vector label is a first steering motor moment coefficient fault and a second steering motor moment coefficient, the first steering motor moment coefficient fault occurs, and the second steering motor moment coefficient is normal; the first steering motor performs corner control through a mu control algorithm, and the second steering motor performs torque control through a PID control algorithm; reminding a driver of replacing the first steering motor in time;
44) if the current fault vector label is that the torque coefficient of the first steering motor is normal and the torque coefficient of the second steering motor is normal, the second steering motor performs corner control through a mu control algorithm, and the first steering motor performs torque control through a PID control algorithm; reminding a driver of replacing the second steering motor in time;
45) if the current fault vector labels are a first steering motor resistance fault and a second steering motor resistance fault, cutting off the current of the first steering motor and the second steering motor, stopping controlling the two steering motors, reminding a driver to stop at the side and timely replacing the first steering motor and the second steering motor;
46) if the current fault vector labels are a first steering motor moment coefficient fault and a second steering motor moment coefficient fault, the control voltages of the first steering motor and the second steering motor are increased, and the two steering motors are controlled through a mu control algorithm; reminding a driver of replacing the first steering motor and the second steering motor in time;
47) if the current fault vector labels are that the first steering motor is all normal and the second steering motor is all normal, the first steering motor and the second steering motor both pass through H2/HinfThe algorithm performs corner control and tracks the ideal front wheel corner.
CN202111646267.4A 2021-12-30 2021-12-30 Dual-motor steer-by-wire system and convolutional neural network fault-tolerant control method thereof Active CN114454951B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111646267.4A CN114454951B (en) 2021-12-30 2021-12-30 Dual-motor steer-by-wire system and convolutional neural network fault-tolerant control method thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111646267.4A CN114454951B (en) 2021-12-30 2021-12-30 Dual-motor steer-by-wire system and convolutional neural network fault-tolerant control method thereof

Publications (2)

Publication Number Publication Date
CN114454951A true CN114454951A (en) 2022-05-10
CN114454951B CN114454951B (en) 2022-12-06

Family

ID=81407253

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111646267.4A Active CN114454951B (en) 2021-12-30 2021-12-30 Dual-motor steer-by-wire system and convolutional neural network fault-tolerant control method thereof

Country Status (1)

Country Link
CN (1) CN114454951B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115951002A (en) * 2023-03-10 2023-04-11 山东省计量科学研究院 Gas chromatography-mass spectrometer fault detection device

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108334814A (en) * 2018-01-11 2018-07-27 浙江工业大学 A kind of AR system gesture identification methods based on convolutional neural networks combination user's habituation behavioural analysis
CN109733464A (en) * 2018-12-18 2019-05-10 南京航空航天大学 Active Fault Tolerant and failure mitigation system and its control method based on steering-by-wire bi-motor
CN110780188A (en) * 2019-10-18 2020-02-11 合肥工业大学 IGBT drive circuit fault diagnosis method
US20200142421A1 (en) * 2018-11-05 2020-05-07 GM Global Technology Operations LLC Method and system for end-to-end learning of control commands for autonomous vehicle
CN112995309A (en) * 2021-02-08 2021-06-18 南京航空航天大学 Internet-connected cloud-control intelligent line-control chassis control system and control method thereof
CN112977602A (en) * 2021-02-04 2021-06-18 南京航空航天大学 Dual-motor steer-by-wire system and hybrid robust stability control method thereof
CN113184040A (en) * 2021-06-03 2021-07-30 长安大学 Unmanned vehicle line-controlled steering control method and system based on steering intention of driver

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108334814A (en) * 2018-01-11 2018-07-27 浙江工业大学 A kind of AR system gesture identification methods based on convolutional neural networks combination user's habituation behavioural analysis
US20200142421A1 (en) * 2018-11-05 2020-05-07 GM Global Technology Operations LLC Method and system for end-to-end learning of control commands for autonomous vehicle
CN109733464A (en) * 2018-12-18 2019-05-10 南京航空航天大学 Active Fault Tolerant and failure mitigation system and its control method based on steering-by-wire bi-motor
CN110780188A (en) * 2019-10-18 2020-02-11 合肥工业大学 IGBT drive circuit fault diagnosis method
CN112977602A (en) * 2021-02-04 2021-06-18 南京航空航天大学 Dual-motor steer-by-wire system and hybrid robust stability control method thereof
CN112995309A (en) * 2021-02-08 2021-06-18 南京航空航天大学 Internet-connected cloud-control intelligent line-control chassis control system and control method thereof
CN113184040A (en) * 2021-06-03 2021-07-30 长安大学 Unmanned vehicle line-controlled steering control method and system based on steering intention of driver

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115951002A (en) * 2023-03-10 2023-04-11 山东省计量科学研究院 Gas chromatography-mass spectrometer fault detection device

Also Published As

Publication number Publication date
CN114454951B (en) 2022-12-06

Similar Documents

Publication Publication Date Title
EP1965993B1 (en) Method and system for assisting a driver when parking or manoeuvring a motor vehicle
DE102016103637B4 (en) PARKING ASSISTANCE TAKING TIRE RADIUS INTO ACCOUNT
CN105667577B (en) Wire-controlled steering system and control method with sensor signal fault tolerance
US6678596B2 (en) Generating steering feel for steer-by-wire systems
CN112519793B (en) Intelligent drive-by-wire chassis architecture based on digital twin and active fault tolerance method thereof
CN105083373B (en) A kind of steering-by-wire road feel device and its control method based on parameter Estimation
DE102016218283A1 (en) AERODYNAMIC SYSTEM AND METHOD FOR DIAGNOSING THE AERODYNAMIC SYSTEM AND VERIFYING THE PRESSURE FORCE ASSESSMENT BASED ON THE ELECTRIC MOTOR CURRENT
CN114454950B (en) Dual-motor steer-by-wire system and fault-tolerant control method thereof
CN112977602B (en) Dual-motor steer-by-wire system and hybrid robust stability control method thereof
CN109733464B (en) Active fault tolerance and fault relief system based on steer-by-wire double motors and control method thereof
CN106467038A (en) The anti-assist steering system turning to failure of electric automobile and assisted diversion method
CN111376971B (en) Rack force-based road surface identification and adaptive steering wheel moment compensation method
CN107140012A (en) A kind of wire-controlled steering system and control method based on the Kalman filter that can suppress diverging
CN108340966A (en) Position self-learning method in electric boosting steering system and its steering wheel angle
CN105774902A (en) Automobile power steering control device with fault-tolerant function and control method
CN114454951B (en) Dual-motor steer-by-wire system and convolutional neural network fault-tolerant control method thereof
CN106882080A (en) A kind of differential steering system and its adaptive neural network fault tolerant control method
CN115339507B (en) Steering control system
CN111422249B (en) Rear wheel steering control method, rear wheel steering control device and computer storage medium
EP4017786B1 (en) Method and device for estimating a steering wheel torque for mechanical feedback at a steering wheel of a steering system of a motor vehicle
CN205381303U (en) Drive -by -wire a steering system with fault -tolerant function of sensor signal
EP3039398B1 (en) Method for the simulation of cornering
CN206871167U (en) A kind of steering-by-wire control system based on parameter Estimation
DE102006060457A1 (en) Driver assisting method for motor vehicle, involves providing vehicle safety system, predicting desired path in calculation model by differences in traveling distance, and detecting surroundings of vehicle by driver assistance system
CN113562066B (en) Wire-controlled steering system and fault diagnosis method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant