A method and system for controlling steering gear ratio of heavy duty transport equipment under a takeover operating condition

Abstract

The application discloses a kind of takeover working condition heavy carrying equipment steering transmission ratio control method and system, comprising: in response to driver input steering wheel angle signal, obtain the psychological related signal and physiological related signal of driver;According to the psychological related signal, determine the emotional state probability vector of driver, according to the physiological related signal, determine the arm muscle fatigue degree of driver;According to the emotional state probability vector and the arm muscle fatigue degree, determine personalized steering transmission ratio;Get current actual angle, according to the personalized steering transmission ratio and angle signal determine target front wheel angle, according to the target front wheel angle and current actual angle control line control steering actuator, the present application is by fusing the change characteristics of psychological and physiological state of driver at takeover moment, realize the dynamic adjustment of steering transmission ratio, to solve the problem that current control strategy is difficult to deal with driver psychological load fluctuation, physiological fatigue and operation lag.

Description

Technical Field

[0001] This invention relates to a method and system for controlling the steering transmission ratio of heavy transport equipment under takeover conditions, belonging to the field of steering technology for heavy transport equipment. Background Technology

[0002] In recent years, with the continuous development of heavy-duty transport equipment towards intelligent and drive-by-wire technologies, the application of unmanned heavy-duty transport equipment in complex scenarios such as mining transportation, port handling, and tactical material delivery has rapidly expanded, placing higher demands on the safety of human-machine co-driving modes. Compared to ordinary passenger vehicles, heavy-duty transport equipment is characterized by its large overall mass, high inertia, slow speed changes, and long steering response cycle. Once the unmanned driving system experiences sensor obstruction, sudden environmental changes, or algorithm uncertainties that trigger emergency takeover, the driver must regain effective control of the vehicle within a very short time; otherwise, they will face serious safety threats.

[0003] In frequent or sudden emergency takeover scenarios, drivers must quickly switch from passive monitoring to active control upon taking over. Their reaction speed, steering force output, and operational smoothness are significantly affected by increased psychological load and fluctuations in physiological state. When drivers are in a state of attention drift, decreased alertness, or physiological fatigue, their steering operations often exhibit significant lag and are prone to excessive or insufficient maneuvering. For heavy-duty transport equipment with high inertia, large steering damping, and low path tracking tolerance, these operational deviations generated by the driver during takeover are further amplified due to their inherent dynamic characteristics. If the steer-by-wire system cannot adaptively adjust the steering ratio according to the driver's psychological and physiological state during takeover, the steering response may be too fast or too slow, thereby exacerbating lateral vehicle drift, causing load sway, and significantly increasing safety risks.

[0004] Existing steering ratio control methods primarily adjust based on vehicle speed and stability indicators, catering mainly to passenger car usage scenarios and assuming the driver is in a good operating state. These methods struggle to adapt to the large inertia dynamics, low response margin, and drastic changes in driver state that accompany the takeover phase of heavy-duty transport equipment. Especially when the psychological switching cost and physiological workload increase dramatically during takeover, steering ratio adjustments based on traditional logic are prone to being too high or too low. This increases the difficulty of trajectory correction for heavy-duty transport equipment, hindering the safety and controllability of the takeover process.

[0005] It is evident that, in order to avoid amplification of lateral offset caused by excessively fast or slow steering response, thereby improving driving comfort and safety during emergency takeover, there is an urgent need for a steering transmission ratio control method and system for heavy-duty transport equipment under takeover conditions. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and system for controlling the steering ratio of heavy transport equipment under takeover conditions. By integrating the changes in the driver's psychological and physiological state at the moment of takeover, the method achieves dynamic adjustment of the steering ratio, thereby solving the problem that current control strategies are unable to cope with fluctuations in the driver's psychological load, physiological fatigue, and operational lag.

[0007] To achieve the above objectives, the present invention is implemented using the following technical solution:

[0008] In a first aspect, the present invention provides a method for controlling the steering transmission ratio of heavy transport equipment under takeover conditions, comprising:

[0009] In response to the driver's input of steering wheel angle signal, acquire the driver's psychological and physiological signals;

[0010] The driver's emotional state probability vector is determined based on the psychologically relevant signals, and the driver's arm muscle fatigue is determined based on the physiologically relevant signals.

[0011] The personalized steering ratio is determined based on the emotional state probability vector and the arm muscle fatigue level.

[0012] The current actual steering angle is obtained, and the target front wheel steering angle is determined based on the personalized steering gear ratio and steering angle signal. The PID control algorithm is used to control the steer-by-wire actuator based on the target front wheel steering angle and the current actual steering angle.

[0013] Furthermore, based on the emotional state probability vector and the arm muscle fatigue level, a personalized steering ratio is determined, including:

[0014] The emotional state probability vector is input into a pre-built psychological channel calculation model to obtain the initial steering gear ratio output by the model.

[0015] The arm muscle fatigue level is input into a pre-built physiological channel calculation model to obtain the physiological correction gain output by the model.

[0016] The initial steering ratio is corrected based on the physiological correction gain to obtain a personalized steering ratio.

[0017] Furthermore, the psychologically relevant signals include facial expression signals, pupil diameter sequence signals, facial temperature signals, and eye movement behavior signals;

[0018] Determining the driver's emotional state probability vector based on the aforementioned psychologically relevant signals includes:

[0019] The psychologically relevant signals are preprocessed to obtain effective signals;

[0020] Feature extraction is performed on the effective signal to obtain facial expression feature vector, infrared feature vector and eye movement feature vector. The facial expression feature vector, infrared feature vector and eye movement feature vector are combined to form a psychological feature matrix. The facial expression feature vector is determined by facial expression signal, the infrared feature vector is determined by pupil diameter sequence signal and facial temperature signal, and the eye movement feature vector is determined by eye movement behavior signal.

[0021] The psychological feature matrix is ​​processed to obtain an effective temporal feature vector;

[0022] The effective temporal feature vector is input into the Long Short-Term Memory network to obtain the hidden state output of the last layer;

[0023] The hidden state output is input into a fully connected layer to obtain an emotional state probability vector.

[0024] Furthermore, the physiologically relevant signals include multi-channel arm muscle signals;

[0025] Determining the driver's arm muscle fatigue level based on the aforementioned physiological signals includes:

[0026] The arm muscle signals from each channel were preprocessed to obtain effective electromyographic signals;

[0027] The effective electromyographic signals were analyzed in the time and frequency domains to extract the root mean square value and average frequency of each channel.

[0028] The root mean square value and average frequency of each channel are weighted and fused to obtain the overall root mean square value and overall average frequency.

[0029] Arm muscle fatigue is determined based on the overall root mean square value and the overall average frequency.

[0030] Furthermore, the formula for calculating the initial steering gear ratio is as follows:

[0031] ;

[0032] In the formula, As the initial steering gear ratio, the emotional state probability vector includes confidence levels corresponding to calm state, mild tension state, severe tension state, and fatigue state. This is the first steering gear ratio. Confidence level for the calm state. This is the second steering gear ratio. The confidence level is for a state of mild stress. This is the third steering gear ratio. The confidence level for a state of severe stress. This is the fourth steering gear ratio. For the confidence level of fatigue state, the first Confidence level corresponding to class state , ;

[0033] The formula for calculating the physiological correction gain is as follows:

[0034] ;

[0035] ;

[0036] In the formula, For physiological correction gain, For arm muscle fatigue, This is the fatigue adjustment function. It is a natural constant. It is a physiological correction factor;

[0037] The formula for calculating the personalized steering ratio is as follows:

[0038] ;

[0039] In the formula, For personalized steering ratios.

[0040] Furthermore, the step of controlling the steer-by-wire actuator using a PID control algorithm based on the target front wheel steering angle and the current actual steering angle includes:

[0041] ;

[0042] In the formula, For the control quantity of the steering actuator by wire, , , These are the proportional coefficient, integral coefficient, and differential coefficient, respectively. This is for angular error;

[0043] The formula for calculating the angle error is as follows:

[0044] ;

[0045] ;

[0046] In the formula, For the target front wheel steering angle, This is the current actual turning angle. This is the steering wheel angle signal.

[0047] Furthermore, the formula for calculating the emotional state probability vector is as follows:

[0048] ;

[0049] In the formula, This is the weight matrix of the fully connected layer. This is the bias vector for the fully connected layer. Output the hidden state of the last layer. This is the operator for the normalization exponential function;

[0050] The formula for calculating the output of the hidden state of the last layer is as follows:

[0051] ;

[0052] In the formula, , , Each is the current number The activation values ​​of the input gate, forget gate, and output gate at each sampling point. For the current number Candidate memory states of each sampling point For the current number The memory unit state of each sampling point For the current number The hidden state output of each sampling point For the previous moment The hidden state output of each sampling point For the previous moment The memory unit state of each sampling point , , , These are the weight matrices corresponding to the input gate, forget gate, output gate, and candidate memory states, respectively. , , , These are the weight matrices of the hidden state from the previous time step corresponding to the input gate, forget gate, output gate, and candidate memory state, respectively. , , , These are the bias terms corresponding to the input gate, forget gate, output gate, and candidate memory states, respectively. It is the hyperbolic tangent function. For Hadamard multiplication, For the first Effective temporal feature vectors of each sampling point It is a non-linear activation function.

[0053] Secondly, the present invention provides a steering transmission ratio control system for heavy transport equipment under takeover conditions, comprising:

[0054] The first module is used to respond to the driver's input of the steering wheel angle signal and to acquire the driver's psychological and physiological signals.

[0055] The second module is used to determine the driver's emotional state probability vector based on the psychologically relevant signals and to determine the driver's arm muscle fatigue based on the physiologically relevant signals.

[0056] The third module is used to determine the personalized steering ratio based on the emotional state probability vector and the arm muscle fatigue.

[0057] The fourth module is used to obtain the current actual steering angle, determine the target front wheel steering angle based on the personalized steering transmission ratio and steering angle signal, and use a PID control algorithm to control the steer-by-wire actuator based on the target front wheel steering angle and the current actual steering angle.

[0058] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described above.

[0059] Fourthly, the present invention provides a computer device, comprising:

[0060] Memory, used to store computer programs / instructions;

[0061] A processor for executing the computer program / instructions to implement the steps of any of the methods described above.

[0062] Fifthly, the present invention provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of any of the methods described above.

[0063] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0064] This invention provides a method and system for controlling the steering ratio of heavy transport equipment under takeover conditions. By integrating the changes in the driver's psychological and physiological state at the moment of takeover, the steering ratio is dynamically adjusted, effectively avoiding the amplification of lateral offset caused by excessively fast or slow steering response, thereby improving driving comfort and driving safety during emergency takeover.

[0065] This invention provides a method and system for controlling the steering ratio of heavy transport equipment under takeover conditions. When the system detects driver stress or increased cognitive load, it automatically reduces the steering ratio through a psychological channel to suppress over-operation, reduce the amplification effect of lateral offset, and improve the directional stability of the heavy transport equipment. When the system detects physiological changes such as increased arm muscle fatigue and decreased steering force output, it increases the steering ratio through a physiological channel to reduce the operating load and improve the smoothness of operation during takeover.

[0066] This invention provides a method and system for controlling the steering transmission ratio of heavy transport equipment under takeover conditions. Through a series adaptive adjustment mechanism with psychological and physiological dual channels, it can maintain a dynamic balance between handling sensitivity and control stability under different driver conditions, significantly improving the handling comfort and driving safety of unmanned heavy transport equipment in emergency takeover conditions. Attached Figure Description

[0067] Figure 1 This is a flowchart illustrating a method for controlling the steering transmission ratio of heavy transport equipment under takeover conditions, as provided in an embodiment of the present invention. Detailed Implementation

[0068] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments and specific features in the embodiments are detailed descriptions of the technical solution of the present application, rather than limitations thereof. In the absence of conflict, the embodiments and technical features in the embodiments can be combined with each other.

[0069] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0070] Example 1:

[0071] Figure 1 This is a flowchart of a method for controlling the steering ratio of heavy transport equipment under a takeover operation, as described in Embodiment 1 of the present invention. The steering ratio control method for heavy transport equipment under a takeover operation provided in this embodiment can be applied to a terminal and can be executed by a steering ratio control system for heavy transport equipment under a takeover operation. This system can be implemented in software and / or hardware and can be integrated into the terminal, such as any smartphone, tablet, or computer device with communication capabilities. See also... Figure 1 The method implemented in this way specifically includes the following steps:

[0072] In response to the driver's input of steering wheel angle signal, acquire the driver's psychological and physiological signals;

[0073] The driver's emotional state probability vector is determined based on the psychologically relevant signals, and the driver's arm muscle fatigue is determined based on the physiologically relevant signals.

[0074] The personalized steering ratio is determined based on the emotional state probability vector and the arm muscle fatigue level.

[0075] The current actual steering angle is obtained, and the target front wheel steering angle is determined based on the personalized steering gear ratio and steering angle signal. The PID control algorithm is used to control the steer-by-wire actuator based on the target front wheel steering angle and the current actual steering angle.

[0076] The method for controlling the steering ratio of heavy transport equipment under takeover conditions provided in this embodiment involves the following steps in its application:

[0077] The driver's psychological signals are collected by a multimodal perception unit located in the cockpit of heavy transport equipment. The psychological signals include facial expression image signals, pupil diameter sequence signals, facial temperature signals, and eye movement signals. The multimodal perception unit includes a facial expression recognition camera, an infrared imaging sensor, and an eye movement sensor. The sampling frequency of each sensor is not less than 30Hz, and the timestamps are aligned to ensure synchronous signal acquisition.

[0078] The acquired raw signal is passed through a low-pass filter. To remove high-frequency noise, the cutoff frequency was set to 5Hz, resulting in the filtered signal. :

[0079]

[0080] In the formula, For a moment The original signal, For a moment The effective signal after filtering.

[0081] Feature extraction is performed on the filtered effective signal to obtain facial expression, infrared, and eye movement feature vectors, which are then combined to form a psychological feature matrix.

[0082] First, the facial expression image signal is divided into The gray-level difference change of each unit is calculated to characterize the intensity of facial expression activation. Facial activation intensity per unit Defined as:

[0083]

[0084] In the formula, For a moment The current unit image of the first Each pixel's grayscale value The average grayscale value of the pixels. This represents the number of pixels per unit.

[0085] This yields the facial expression feature vector. :

[0086]

[0087] Subsequently, the pupil diameter change rate was calculated using the pupil diameter sequence signal. :

[0088]

[0089] In the formula, and They represent Time and Pupil diameter sequence signal at time t.

[0090] The average facial temperature of the driver was extracted based on facial temperature signals. Calculate the change in facial temperature :

[0091]

[0092] In the formula, The initial temperature of the driver's face.

[0093] This yields the infrared feature vector. :

[0094]

[0095] Finally, blink frequency, average gaze duration, and gaze direction deviation were obtained based on eye movement signals. Among these, blink frequency... The calculation is as follows:

[0096]

[0097] In the formula, Time period Number of internal blinks.

[0098] Average gaze duration for:

[0099]

[0100] In the formula, For the first Duration of the second gaze Time period Number of inward gazes.

[0101] gaze direction deflection angle The calculation expression is:

[0102]

[0103] In the formula, Let be the vector representing the driver's line of sight. Let be the vector representing the vehicle's forward direction. The magnitude of the line-of-sight vector. The magnitude of the vector representing the vehicle's forward direction. It is an inverse cosine function.

[0104] This yields the eye-tracking feature vector. :

[0105]

[0106] The above facial expression, infrared, and eye movement feature vectors are combined to form a psychological feature matrix. :

[0107]

[0108] A multimodal emotion recognition model based on the fusion of convolutional neural networks and long short-term memory networks is constructed. The model input is a psychological feature matrix, and the model output is a probability vector of the driver's emotional state, namely calm state, mild tension state, severe tension state, and fatigue state. The specific recognition steps are as follows:

[0109] First, the psychological feature matrix is ​​processed by synchronously sampling and fusing each feature vector along the time dimension to obtain the temporal unfolded form:

[0110]

[0111] In the formula, Indicates the first The temporal unfolded feature vector of each sampling point , This represents the number of sampling points.

[0112] The input matrix is ​​processed by convolution. Perform local feature extraction to obtain the first Intermediate features of each sampling point Represented as:

[0113]

[0114] In the formula, The convolution kernel weight matrix is... For the corresponding bias term, This represents a one-dimensional convolution operation. It is a non-linear activation function.

[0115] The intermediate features are obtained by dimensionality reduction using a pooling layer:

[0116]

[0117] In the formula, The length of the pooling layer window. This is the effective temporal feature vector after pooling. No. Intermediate features of each sampling point This is the offset index within the pooling window. This is a maximum pooling operation.

[0118] Subsequently, the effective temporal feature sequence is input into the Long Short-Term Memory network to model the time dependency relationship. The state update equation is as follows:

[0119]

[0120] In the formula, , , Each is the current number The activation values ​​of the input gate, forget gate, and output gate at each sampling point. For the current number Candidate memory states of each sampling point For the current number The memory cell state of each sampling point For the current number The hidden state output of each sampling point For the previous moment The hidden state output of each sampling point For the previous moment The memory cell state of each sampling point , , , These are the weight matrices corresponding to the input gate, forget gate, output gate, and candidate memory states, respectively. , , , These are the weight matrices of the hidden state from the previous time step corresponding to the input gate, forget gate, output gate, and candidate memory state, respectively. , , , These are the bias terms corresponding to the input gate, forget gate, output gate, and candidate memory states, respectively. It is the hyperbolic tangent function. For Hadamard multiplication, For the first Each sampling point has an effective temporal feature vector.

[0121] The hidden state vector from the last layer is input into the fully connected layer and normalized using the Softmax function to obtain the emotional state probability vector. :

[0122]

[0123] In the formula, This is the weight matrix of the fully connected layer. For the bias vector of the fully connected layer, For normalized exponential function operators, , , , The confidence levels are respectively for calm state, mild stress state, severe stress state, and fatigue state, and the [number]th [state] is [value]. Confidence level corresponding to class state , A multi-channel surface electromyography (EMG) sensor unit, deployed on the driver's forearm, collects real-time muscle signals from the driver's arm during driving. The multi-channel EMG sensor unit includes differential electrodes, an amplification and filtering module, and an analog-to-digital converter module. It is used to collect EMG signals from the driver's arm muscles, with the acquisition time... raw electromyographic signals Recorded as:

[0124]

[0125] In the formula, The total number of channels. For the first Channel at time electromyographic signals, .

[0126] The acquired raw electromyography (EMG) signals were preprocessed using a bandpass filter. Remove high-frequency noise to obtain the time. Effective electromyographic signals It is expressed by the following formula:

[0127]

[0128] In the formula, Its passband frequency range is 20–450 Hz, which is used to preserve the frequency band of major muscle activity.

[0129] Time-domain and frequency-domain analyses were performed on the filtered effective electromyography signals to extract key feature parameters that reflect changes in driver's arm muscle fatigue, including the root mean square value in the time domain and the average frequency in the frequency domain.

[0130] No. Root mean square value of the channel The formula used to measure the overall strength of muscle contraction is as follows:

[0131]

[0132] In the formula, Indicates the filtered first... Channel at time electromyographic signals, The sampling period.

[0133] No. average frequency of the channel The formula used to reflect the central position of the effective electromyographic signal in the frequency distribution is as follows:

[0134]

[0135] In the formula, The filtered first Channel electromyography signals at frequency Power spectral density at that point This is the maximum frequency.

[0136] The root mean square (RMS) values ​​of each channel are weighted and fused with the average frequency to obtain the overall RMS value of the electromyographic signal. With average frequency The specific calculations are as follows:

[0137]

[0138]

[0139] In the formula, For the first The weighting coefficients of the channel, and satisfying .

[0140] Calculate driver's arm muscle fatigue This is used to continuously quantify the fatigue level of a driver's arm. During muscle fatigue, the root mean square value increases with the degree of muscle fatigue, while the average frequency decreases as fatigue deepens. To achieve a smooth and monotonic normalized mapping, the expression for arm muscle fatigue is defined based on the sigmoid function as follows:

[0141]

[0142] In the formula, These are the weighting coefficients. , These are the slope coefficients corresponding to the root mean square value and the average frequency, respectively, used to adjust the sensitivity to changes. and These are the root mean square value and the average frequency, respectively, as reference values ​​for the driver's initial state. and These are the shift parameters corresponding to the root mean square value and the average frequency, respectively. , , It is a natural constant.

[0143] Define steering ratio ,in For the front wheel steering angle, This refers to the steering wheel angle. A higher gear ratio indicates higher steering sensitivity and lower operating load; a lower gear ratio indicates smoother steering response and higher system stability.

[0144] A psychological channel calculation model is constructed, taking the driver's emotional state recognition results as input, considering the differences in control needs corresponding to different emotional states, and using fuzzy reasoning logic to determine the steering transmission ratio.

[0145] First, establish the following fuzzy rules:

[0146] Rule R1: If the driver is calm, the steering ratio is taken as the reference value, i.e., the first steering ratio. , This corresponds to normal steering sensitivity;

[0147] Rule R2: If the driver is in a slightly tense state, the steering ratio should be slightly smaller, i.e., the second steering ratio. , , As the first variable, it is used to suppress steering wheel angle fluctuations and oversteering caused by tension;

[0148] Rule R3: If the driver is in a state of severe tension, the steering ratio shall be a smaller value, i.e., the third steering ratio. , , As a second variable, it further reduces steering sensitivity to prevent vehicle instability caused by accidental touch or sudden steering.

[0149] Rule R4: If the driver is fatigued, the steering ratio should be the larger value, i.e., the fourth steering ratio. , , As a third variable, it is used to reduce the driver's workload and compensate for the steering delay caused by slow response.

[0150] in, The reference gear ratio calibrated for the steering system. , .

[0151] The emotional state probability vector output by the Softmax function As a weighting coefficient matrix, the steering ratio of the psychological channel output is calculated. :

[0152]

[0153] Since the probability vector of emotional state reflects the confidence distribution of the driver in each emotional state, in order to avoid sudden changes in the steering ratio caused by a single emotional state judgment, this embodiment uses the steering ratio corresponding to each rule as the rule output and performs weighted fusion with the corresponding emotional state confidence as the weight.

[0154] A physiological pathway calculation model is constructed, using the arm muscle fatigue calculation results as input, and a fatigue regulation function based on the Sigmoid function is built. :

[0155]

[0156] In the formula, It is a natural constant.

[0157] Considering the different steering sensitivity requirements of drivers under different levels of fatigue, a physiological correction gain is defined. :

[0158]

[0159] In the formula, This is a physiological correction factor used to adjust the intensity of the effect of fatigue on steering sensitivity. During the operation of heavy-duty transport equipment, as the driver's arm muscle fatigue increases, the physiological correction gain decreases, thereby reducing the driving and control load and improving driving comfort.

[0160] The steering ratio output from the psychological channel is adjusted by using the correction gain calculated through the physiological channel, forming a personalized steering ratio model with a psychological-physiological dual-channel series connection. The final personalized steering ratio... The calculation expression is as follows:

[0161]

[0162] The above formula realizes the dynamic adaptive adjustment of the steering transmission ratio of unmanned heavy transport equipment during the takeover process. It also takes into account the driver's psychological state and physiological load factors. The system can adaptively adjust the balance between steering sensitivity and operating load based on its real-time psychological-physiological characteristics, thereby ensuring the operational stability and comfort of unmanned heavy transport equipment during the takeover process.

[0163] Based on the time entered by the driver Steering wheel angle signal The time of calculation of the personalized transmission ratio obtained Target front wheel steering angle :

[0164]

[0165] Based on the target front wheel steering angle and time Current actual front wheel steering angle Calculation time angular error :

[0166]

[0167] The PID control algorithm is used to perform closed-loop control on the steer-by-wire actuator. The control law is as follows:

[0168]

[0169] In the formula, For a moment The amount of control by the executing agency, , , These are the proportional coefficient, integral coefficient, and derivative coefficient, respectively. The above control quantities are applied to the steering actuator to achieve the target front wheel steering angle output.

[0170] Example 2:

[0171] This embodiment provides a steering transmission ratio control system for heavy transport equipment under takeover conditions, including:

[0172] The first module is used to respond to the driver's input of the steering wheel angle signal and to acquire the driver's psychological and physiological signals.

[0173] The second module is used to determine the driver's emotional state probability vector based on the psychologically relevant signals and to determine the driver's arm muscle fatigue based on the physiologically relevant signals.

[0174] The third module is used to determine the personalized steering ratio based on the emotional state probability vector and the arm muscle fatigue.

[0175] The fourth module is used to obtain the current actual steering angle, determine the target front wheel steering angle based on the personalized steering transmission ratio and steering angle signal, and use a PID control algorithm to control the steer-by-wire actuator based on the target front wheel steering angle and the current actual steering angle.

[0176] The specific functions of each module described above are explained in the relevant content of the method in Embodiment 1, and will not be repeated here.

[0177] Example 3:

[0178] This embodiment provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in Embodiment 1.

[0179] Example 4:

[0180] This embodiment provides a computer device, including:

[0181] Memory, used to store computer programs / instructions;

[0182] A processor for executing the computer program / instructions to implement the steps of the method described in Embodiment 1.

[0183] Example 5:

[0184] This embodiment provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the method described in Embodiment 1.

[0185] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

[0186] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0187] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0188] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0189] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0190] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this disclosure and not to limit its protection scope. Although this disclosure has been described in detail with reference to the above embodiments, those skilled in the art should understand that after reading this disclosure, they can still make various changes, modifications or equivalent substitutions to the specific implementation of the invention, but these changes, modifications or equivalent substitutions are all within the protection scope of the pending claims.

Claims

1. A method for controlling the steering transmission ratio of heavy transport equipment under takeover conditions, characterized in that, include: In response to the driver's input of steering wheel angle signal, acquire the driver's psychological and physiological signals; The driver's emotional state probability vector is determined based on the psychologically relevant signals, and the driver's arm muscle fatigue is determined based on the physiologically relevant signals. The personalized steering ratio is determined based on the emotional state probability vector and the arm muscle fatigue level. The current actual steering angle is obtained, and the target front wheel steering angle is determined based on the personalized steering gear ratio and steering angle signal. The PID control algorithm is used to control the steer-by-wire actuator based on the target front wheel steering angle and the current actual steering angle.

2. The method for controlling the steering transmission ratio of heavy transport equipment under the takeover condition according to claim 1, characterized in that, Based on the emotional state probability vector and the arm muscle fatigue level, a personalized steering ratio is determined, including: The emotional state probability vector is input into a pre-built psychological channel calculation model to obtain the initial steering gear ratio output by the model. The arm muscle fatigue level is input into a pre-built physiological channel calculation model to obtain the physiological correction gain output by the model. The initial steering ratio is corrected based on the physiological correction gain to obtain a personalized steering ratio.

3. The method for controlling the steering transmission ratio of heavy transport equipment under the takeover condition according to claim 2, characterized in that, The psychologically relevant signals include facial expression signals, pupil diameter sequence signals, facial temperature signals, and eye movement behavior signals; Determining the driver's emotional state probability vector based on the aforementioned psychologically relevant signals includes: The psychologically relevant signals are preprocessed to obtain effective signals; Feature extraction is performed on the effective signal to obtain facial expression feature vector, infrared feature vector and eye movement feature vector. The facial expression feature vector, infrared feature vector and eye movement feature vector are combined to form a psychological feature matrix. The facial expression feature vector is determined by facial expression signal, the infrared feature vector is determined by pupil diameter sequence signal and facial temperature signal, and the eye movement feature vector is determined by eye movement behavior signal. The psychological feature matrix is ​​processed to obtain an effective temporal feature vector; The effective temporal feature vector is input into the Long Short-Term Memory network to obtain the hidden state output of the last layer; The hidden state output is input into a fully connected layer to obtain an emotional state probability vector.

4. The method for controlling the steering transmission ratio of heavy transport equipment under the takeover condition according to claim 3, characterized in that, The physiologically relevant signals include multi-channel arm muscle signals; Determining the driver's arm muscle fatigue level based on the aforementioned physiological signals includes: The arm muscle signals from each channel were preprocessed to obtain effective electromyographic signals; The effective electromyographic signals were analyzed in the time and frequency domains to extract the root mean square value and average frequency of each channel. The root mean square value and average frequency of each channel are weighted and fused to obtain the overall root mean square value and overall average frequency. Arm muscle fatigue is determined based on the overall root mean square value and the overall average frequency.

5. The method for controlling the steering transmission ratio of heavy transport equipment under the takeover condition according to claim 4, characterized in that, The formula for calculating the initial steering gear ratio is as follows: ； In the formula, As the initial steering gear ratio, the emotional state probability vector includes confidence levels corresponding to calm state, mild tension state, severe tension state, and fatigue state. This is the first steering gear ratio. Confidence level for the calm state. This is the second steering gear ratio. The confidence level is for a state of mild stress. This is the third steering gear ratio. The confidence level for a state of severe stress. This is the fourth steering gear ratio. For the confidence level of fatigue state, the first Confidence level corresponding to class state , ; The formula for calculating the physiological correction gain is as follows: ； ； In the formula, For physiological correction gain, For arm muscle fatigue, This is the fatigue adjustment function. It is a natural constant. As a physiological correction factor, ; The formula for calculating the personalized steering ratio is as follows: ； In the formula, For personalized steering ratios.

6. The method for controlling the steering transmission ratio of heavy transport equipment under the takeover condition according to claim 5, characterized in that, The steer-by-wire actuator is controlled using a PID control algorithm based on the target front wheel steering angle and the current actual steering angle, including: ； In the formula, for The control quantity of the steering actuator at any given moment. , , These are the proportional coefficient, integral coefficient, and differential coefficient, respectively. for angular error at any given time; The formula for calculating the angle error is as follows: ； ； In the formula, for The target front wheel angle at any given moment for The current actual turning angle at any given moment. for The steering wheel angle signal at any given moment.

7. The method for controlling the steering transmission ratio of heavy transport equipment under the takeover condition according to claim 5, characterized in that, The formula for calculating the probability vector of the emotional state is as follows: ； In the formula, Let be the probability vector of emotional state. This is the weight matrix of the fully connected layer. This is the bias vector for the fully connected layer. Output the hidden state of the last layer. This is the operator for the normalization exponential function; The formula for calculating the output of the hidden state of the last layer is as follows: ； In the formula, , , Each is the current number The activation values ​​of the input gate, forget gate, and output gate at each sampling point. For the current number Candidate memory states of each sampling point For the current number The memory unit state of each sampling point For the current number The hidden state output of each sampling point For the previous moment The hidden state output of each sampling point For the previous moment The memory unit state of each sampling point , , , These are the weight matrices corresponding to the input gate, forget gate, output gate, and candidate memory states, respectively. , , , These are the weight matrices of the hidden state from the previous time step corresponding to the input gate, forget gate, output gate, and candidate memory state, respectively. , , , These are the bias terms corresponding to the input gate, forget gate, output gate, and candidate memory states, respectively. It is the hyperbolic tangent function. For Hadamard multiplication, For the first Each sampling point has an effective temporal feature vector. It is a non-linear activation function.

8. A steering transmission ratio control system for heavy-duty transport equipment under takeover conditions, characterized in that, include: The first module is used to respond to the driver's input of the steering wheel angle signal and to acquire the driver's psychological and physiological signals. The second module is used to determine the driver's emotional state probability vector based on the psychologically relevant signals and to determine the driver's arm muscle fatigue based on the physiologically relevant signals. The third module is used to determine the personalized steering ratio based on the emotional state probability vector and the arm muscle fatigue. The fourth module is used to obtain the current actual steering angle, determine the target front wheel steering angle based on the personalized steering transmission ratio and steering angle signal, and use a PID control algorithm to control the steer-by-wire actuator based on the target front wheel steering angle and the current actual steering angle.

9. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1 to 7.

10. A computer device, characterized in that, include: Memory, used to store computer programs / instructions; A processor for executing the computer program / instructions to implement the steps of the method according to any one of claims 1 to 7.

Images