Method for dynamically adjusting control strategy of remote control equipment based on man-machine trust
By obtaining the multi-channel physiological signals of the operator and using the bidirectional long and short-term memory network with a hybrid attention mechanism to calculate the trust level, dynamically adjusting the control strategy of the remote control device, the problem of insufficient real-time monitoring and dynamic adjustment of emotional state in the existing technology is solved. Improves the operating stability and safety of the remote control system.
Patent Information
- Application Number
- CN202410975033.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-19
- Publication Date
- 2024-12-06
AI Technical Summary
The existing technology has insufficient real-time, accuracy and dynamicity in real-time monitoring and dynamic adjustment of control strategies for remote-controlled unmanned vehicles and collaborative robot operators, resulting in possible misoperation in emergencies, reducing the stability of the system's operation. Sex and safety.
By obtaining multi-channel physiological signals of the operator (ECG, electrocutaneous skin activity, breathing and pulse signals), the operator's trust level is calculated using a bidirectional long and short-term memory network based on a hybrid attention mechanism, and dynamically adjusting the remote control device's trust level according to the trust level Control strategies, including travel speed, steering angle, brake force and acceleration and deceleration rate.
Real-time monitoring and dynamic adjustment of the emotional state of the operator is realized, the operation stability and safety of the human-computer interaction system is improved, the risk of misoperation is significantly reduced, and the work efficiency and operation accuracy are improved.
Smart Images

Figure CN119088083A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of human-computer interaction, and in particular to a method for dynamically adjusting a control strategy of a remote control device based on human-computer trust. Background Art
[0002] With the rapid development of science and technology, remote-controlled unmanned vehicles and collaborative robots are playing an increasingly important role in disaster relief, industrial production, medical care and other emergency scenarios. They are able to enter dangerous areas that are difficult or impossible for humans to reach directly, and perform tasks such as rescue, reconnaissance and material transportation. However, the effective operation of the remote control system is highly dependent on a stable and reliable human-computer interaction mechanism, especially in a dynamic and unpredictable environment, where the operator's status has a crucial impact on the system's operational performance and safety.
[0003] Although modern technology is able to detect obstacles within a certain range, and automatic obstacle avoidance and path planning technologies are becoming increasingly mature, operators often face tremendous psychological pressure in emergency situations, and their instinctive reactions are easily affected by external interference. Emotional states such as tension and panic may lead to wrong judgments and misoperations, which in turn lead to safety accidents. For example, in disaster relief, operators may experience emotional fluctuations due to sudden sound interference or emergencies, and may misoperate unmanned vehicles or robots, resulting in rescue failures or further dangers.
[0004] Existing research mainly focuses on the following aspects:
[0005] 1. Physiological signal monitoring technology: Use physiological signals such as electrocardiogram (ECG), skin electrodermal activity (EDA), respiration and pulse to assess the operator's physiological and psychological state. These technologies have been widely used in the medical and mental health fields to detect patients' emotions and stress levels.
[0006] 2. Emotion recognition algorithm: Based on machine learning and deep learning technology, it extracts features from multi-dimensional physiological signals to identify emotional states. Commonly used models include support vector machine (SVM), random forest (RF), recurrent neural network (RNN) and long short-term memory network (LSTM).
[0007] 3. Output control technology: Although traditional control methods such as proportional-integral-derivative (PID) control, fuzzy control and rule-based systems can improve system stability to a certain extent, they are often unable to respond to operator status changes in a timely manner when faced with complex and changing operating environments and emergency response situations, and are prone to control errors.
[0008] However, the existing technology still has significant deficiencies in the following aspects:
[0009] 1. Real-time: Many emotion recognition systems are unable to monitor and identify the operator’s emotional state in real time, resulting in the inability to adjust the control strategy in a timely manner.
[0010] 2. Accuracy: Existing emotion recognition algorithms lack accuracy and robustness in complex environments and are easily affected by noise and interference.
[0011] 3. Dynamicity: There is a lack of dynamic adjustment mechanism based on the operator's status, and it is impossible to adjust the system's control strategy according to the operator's real-time status in an emergency to reduce the risk of misoperation. Summary of the invention
[0012] The present invention provides a method for dynamically adjusting a control strategy of a remote control device based on human-machine trust, so as to at least solve the above-mentioned technical problems existing in the prior art to a certain extent.
[0013] In order to solve the above technical problems, the present invention provides the following technical solutions:
[0014] On the one hand, the present invention provides a method for dynamically adjusting a control strategy of a remote control device based on human-machine trust, the method comprising:
[0015] Acquire multi-channel physiological signals of operators of remote control equipment in working state;
[0016] Based on the multi-channel physiological signals of the operator, calculate the trust level of the current remote control device to the operator;
[0017] Based on the current trust level of the remote control device in the operator, the control strategy of the remote control device is dynamically adjusted.
[0018] Furthermore, the multi-channel physiological signal of the operator includes: an electrocardiogram signal of the operator, an electrical skin activity signal of the operator, a breathing signal of the operator and a pulse signal of the operator.
[0019] Furthermore, the calculating of the trust level of the current remote control device to the operator based on the multi-channel physiological signal of the operator includes:
[0020] Preprocessing the acquired multi-channel physiological signals of the operator; wherein the preprocessing includes: signal denoising, baseline correction and standardization;
[0021] The pre-processed multi-channel physiological signals of the operator are input into a bidirectional long short-term memory network based on a hybrid attention mechanism to obtain the operator's emotion type and emotional arousal;
[0022] Based on the emotion type and emotional arousal of the current operator, the trust level of the current remote control device in the operator is calculated.
[0023] Furthermore, the preprocessing of the acquired multi-channel physiological signals of the operator includes:
[0024] The electrocardiogram signal is filtered by a bandpass filter, the baseline drift of the filtered electrocardiogram signal is removed by wavelet transform, the baseline correction of the electrocardiogram signal is realized, and the zero mean normalization algorithm is used to standardize the electrocardiogram signal after the baseline correction;
[0025] The skin electrical activity signal is filtered by a low-pass filter, the baseline drift of the filtered skin electrical activity signal is removed by wavelet transform, the baseline correction of the skin electrical activity signal is achieved, and the baseline-corrected skin electrical activity signal is standardized by a zero-mean normalization algorithm;
[0026] A low-pass filter is used to filter the respiratory signal, a high-pass filter is used to process the low-frequency baseline drift of the filtered respiratory signal, and a baseline correction of the respiratory signal is achieved. A zero-mean normalization algorithm is used to standardize the respiratory signal after baseline correction;
[0027] A low-pass filter is used to filter the pulse signal, and a high-pass filter is used to process the low-frequency baseline drift of the filtered pulse signal to achieve baseline correction of the pulse signal. A zero-mean normalization algorithm is used to standardize the pulse signal after baseline correction.
[0028] Furthermore, the bidirectional long short-term memory network based on the hybrid attention mechanism includes: an input layer, a Bi-LSTM layer, an internal attention layer, an external attention layer and an output layer;
[0029] The process of processing input data by the bidirectional long short-term memory network based on the hybrid attention mechanism includes:
[0030] The Bi-LSTM layer is used to process the time series data inputted by the input layer at each time step to obtain the output of each time step; and the output of each time step is respectively transmitted to the internal attention layer;
[0031] The internal attention layer is used to determine the importance of each time point within each independent signal; the internal attention layer processes the data by first transforming the output of the Bi-LSTM layer through a linear layer and a nonlinear activation function, and then calculating the attention weight of each time step, which is expressed as follows:
[0032] u t =tanh(W c h t +bc )
[0033]
[0034] Among them, W c is a weight matrix used to transform the hidden state to the output h of the Bi-LSTM layer at the current time step t t The attention space of T represents the transpose of the matrix; b c is the bias term; tanh() represents the hyperbolic tangent activation function, which is used to introduce nonlinear transformation; u t' represents the transformation variable at time step t'; t' represents all possible time steps, which are used to calculate the normalized attention weights;
[0035] Then, use α t Get h t The weighted feature representation c is expressed as follows:
[0036] c=∑ t α t h t
[0037] After obtaining the weighted feature representation of a single signal through the internal attention layer, the external attention layer is used to evaluate the contribution of different types of signals to the recognition task; the expression is as follows:
[0038] β i =sigmoid(W s c i +b s )
[0039] s=∑ i β i c i
[0040] Among them, W s and b s is a learning parameter; sigmoid() represents an S-shaped activation function used to map the input to the range of 0 to 1; c i represents the feature representation of the i-th signal;
[0041] Finally, the output layer of the network uses two branches to simultaneously output the emotion type and emotional arousal;
[0042] The branch that outputs the emotion category is used to predict the probability distribution of the emotion category corresponding to the input data. The prediction process is as follows: Assuming that there are N emotion categories, s is mapped to a vector of length N through a fully connected layer, representing the score of each emotion category. Then, the Softmax activation function is used to convert the emotion category score into a probability distribution, so that the probability value of each category is in the range of [0,1], and the sum of the probabilities of all categories is 1;
[0043] The branch that outputs the emotional arousal is used to predict the emotional arousal corresponding to the input data. The prediction process is: s is mapped to a scalar value representing the emotional arousal through a fully connected layer.
[0044] Furthermore, the calculation of the trust level of the current remote control device in the operator based on the emotion type and emotional arousal of the current operator includes:
[0045] The trust score corresponding to the emotion type of the current operator and the trust score corresponding to the emotional arousal degree of the current operator are set based on a preset scoring table; wherein the scoring table includes trust scores corresponding to different types of emotions and trust scores corresponding to different emotional arousal degrees;
[0046] Set the weights of emotion type and emotion arousal on trust score respectively;
[0047] Calculate the trust level of the current remote control device to the operator, the formula is as follows:
[0048] P=W E ×S E +W A ×S A
[0049] Where P is the trust level of the current remote control device to the operator; W E is the influence weight of emotion type on trust score; S E is the trust score corresponding to the emotion type of the current controller; W A is the influence weight of emotional arousal on trust score; S A It is the trust score corresponding to the emotional arousal of the current controller.
[0050] Furthermore, dynamically adjusting the control strategy of the remote control device based on the current trust level of the remote control device to the operator includes:
[0051] The current trust level of the remote control device in the operator is compared with the preset threshold; if the current trust level of the remote control device in the operator is not less than the preset threshold, the operator continues to control the device; otherwise, based on the current trust level of the remote control device in the operator and combined with the individual differences of the operators, a safety adjustment coefficient is calculated, and the operating status of the device is adjusted based on the safety adjustment coefficient.
[0052] Furthermore, the operating state of the device includes: the travel speed of the device, the steering angle of the device, the braking force of the device, and the acceleration and deceleration rates of the device.
[0053] Furthermore, according to the current trust level of the remote control device in the operator and the individual differences of the operators, a safety adjustment coefficient is calculated, and the operating state of the device is adjusted based on the safety adjustment coefficient, including:
[0054] According to the current trust level of the remote control device in the operator, combined with the individual differences of the operators, weights are dynamically allocated. The expressions of each weight are as follows:
[0055] W1=h1(P,S Individual Difference )=a1P+b1
[0056] W2=h2(P,S Individual Difference )=a2P+b2
[0057] W3=h3(P,S Individual Difference )=a3P+b3
[0058] W4=h4(P,S Individual Difference )=a4P+b4
[0059] Based on the assigned weights, the safety adjustment factor is calculated as follows:
[0060] S1=A(Emontion)×W1
[0061] S2=A(Emontion)×W2
[0062] S3=A(Emontion)×W3
[0063] S4=A(Emontion)×W4
[0064] Among them, h1, h2, h3 and h4 are weight functions determined according to the trust level of the current remote control device in the operator and the individual differences of the operators; a1, a2, a3 and a4 are adjustment factors in the weight function that can be set in the range of 0 to 1; b1, b2, b3 and b4 are used to adjust the individual differences S Individual DifferenceThe set bias; W1, W2, W3 and W4 are weight coefficients. By adjusting the weights, different operating environments and user needs can be flexibly responded to; A (Emontion) represents the standardized value of emotional arousal.
[0065] The travel speed of the device is adjusted based on S1; the steering angle of the device is adjusted based on S2; the braking force of the device is adjusted based on S3; and the acceleration and deceleration rates of the device are adjusted based on S4.
[0066] Furthermore, the adjustment formula for the travel speed of the equipment is:
[0067] V'=V·(1-α·S1)
[0068] Wherein, V' represents the travel speed of the device after adjustment; V represents the travel speed of the device before adjustment; α represents the weight coefficient, which can be set to a value of 0 to 1 according to the actual situation;
[0069] The adjustment formula for the steering angle of the device is:
[0070] θ'=θ·(1-β·S2)
[0071] Among them, θ' represents the steering angle of the device after adjustment; θ represents the steering angle of the device before adjustment; β represents the weight coefficient, which can be set to a value of 0 to 1 according to the actual situation;
[0072] The adjustment formula for the braking force of the equipment is:
[0073] F'=F·(1-χ·S3)
[0074] Among them, F' represents the braking force of the device after adjustment; F represents the braking force of the device before adjustment; χ represents the weight coefficient, which can be set to a value of 0 to 1 according to the actual situation;
[0075] The adjustment formula for the acceleration and deceleration rate of the equipment is:
[0076] a'=a·(1-δ·S4)
[0077] Wherein, a' represents the acceleration and deceleration rate of the device after adjustment; a represents the acceleration and deceleration rate of the device before adjustment; δ represents the weight coefficient, which can be set to a value of 0 to 1 according to actual conditions.
[0078] On the other hand, the present invention further provides a remote control device control strategy dynamic adjustment system based on human-machine trust, the remote control device control strategy dynamic adjustment system based on human-machine trust comprising:
[0079] A data acquisition module, used to acquire multi-channel physiological signals of the operator of the remote control device in the working state;
[0080] A remote control device trust level calculation module for the operator, used to calculate the current remote control device trust level for the operator based on the multi-channel physiological signals of the operator acquired by the data acquisition module;
[0081] The control strategy dynamic adjustment module is used to dynamically adjust the control strategy of the remote control device based on the current trust level of the remote control device to the operator calculated by the trust level calculation module of the remote control device to the operator.
[0082] On the other hand, the present invention further provides an electronic device, comprising a processor and a memory; wherein the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the above method.
[0083] In yet another aspect, the present invention further provides a computer-readable storage medium, wherein at least one instruction is stored in the storage medium, and the instruction is loaded and executed by a processor to implement the above method.
[0084] The beneficial effects brought about by the technical solution provided by the present invention include at least:
[0085] 1. The present invention can comprehensively and accurately collect multi-channel physiological signals of remote control operators during their work, which is beneficial to the subsequent design and application of various physiological signal analysis and output control algorithms.
[0086] 2. The present invention can simultaneously monitor and mark emotional states, and automatically adjust the system control strategy according to the current working state. It has the advantages of being wearable, non-invasive, reliable in information processing, and highly automated, and can significantly improve the operational stability and safety of the human-computer interaction system.
[0087] 3. The present invention adopts the attention mechanism and long short-term memory network to recognize the emotional state, and uses four physiological signals, namely electrocardiogram, skin electrical response, breathing signal and pulse signal, for learning, which has the advantages of high information utilization and accurate recognition results.
[0088] 4. The present invention has a high level of intelligence and flexibility. The control output algorithm can be adjusted in real time according to the current state, and corresponding control strategies can be given according to the different states of different operators. Experimental results show that this method significantly improves work efficiency and operation accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0089] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0090] Figure 1 It is a schematic diagram of the execution flow of a method for dynamically adjusting a remote control device control strategy based on human-machine trust provided by an embodiment of the present invention;
[0091] Figure 2 is a flow chart of multi-channel physiological signal preprocessing provided by an embodiment of the present invention;
[0092] Figure 3 It is a structural diagram of a bidirectional long short-term memory network model based on a hybrid attention mechanism provided by an embodiment of the present invention;
[0093] Figure 4 It is a schematic diagram of the working process of the user working state driving control method provided by an embodiment of the present invention;
[0094] Figure 5 is a block diagram of a remote control device control strategy dynamic adjustment system based on human-machine trust provided by an embodiment of the present invention;
[0095] Figure 6 It is a system block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0096] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.
[0097] First of all, it should be noted that in the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "exemplarily" is intended to present the concept in a concrete way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.
[0098] First embodiment
[0099] In order to accurately capture and quantify the level of trust that machines have in humans, and to implement dynamic adjustments based on trust in real-time control systems, this embodiment provides a method for dynamically adjusting the control strategy of a remote control device based on human-machine trust, so as to improve the operational stability and safety of a remote-controlled robot or vehicle in an emergency. The method can be implemented by an electronic device, and the execution process of the method is as follows: Figure 1 As shown, the following steps are included:
[0100] S1, obtaining multi-channel physiological signals of the operator of the remote control device in the working state;
[0101] In this embodiment, the multi-channel physiological signal of the operator includes: an electrocardiogram (ECG) signal of the operator, an electrodermal activity (EDA) signal of the operator, a breathing signal of the operator, and a pulse signal of the operator.
[0102] In a feasible implementation manner, the physiological signal acquisition device may include a collector, multiple sensors, Bluetooth, a marker, a sensor fixing belt and other accessories.
[0103] S2, based on the multi-channel physiological signals of the operator, calculating the trust level of the current remote control device to the operator;
[0104] Specifically, in this embodiment, the calculation process of the trust level of the current remote control device to the operator is as follows:
[0105] S21, preprocessing the acquired multi-channel physiological signals of the operator; wherein the preprocessing includes: signal denoising, baseline correction and standardization; the specific preprocessing process is as follows Figure 2 As shown, including:
[0106] S211, performing signal denoising on physiological signals;
[0107] In a feasible implementation, following the generation characteristics and change laws of various physiological signals, a bandpass filter is used to filter the ECG signal to remove high-frequency and low-frequency noise, such as power frequency interference and electromyographic noise. For EDA signals, high-frequency noise is removed by using a low-pass filter. Respiration and pulse frequency signals are low-pass filtered to remove high-frequency interference and maintain the basic rhythm of breathing and heartbeat. The above denoising is to ensure the accurate transmission of key physiological information. The formula of the bandpass filter can be expressed as:
[0108] H bp (f) = H hp (f) H lp (f) (1)
[0109] Where f is the frequency, H bp(f) is the frequency response of the bandpass filter, which can be expressed as a high-pass filter H hp (f) and a low-pass filter H lp (f) combination.
[0110] S212, performing baseline correction on the denoised physiological signal;
[0111] In a feasible implementation, wavelet transform is used to remove the baseline drift of ECG and EDA signals caused by long-term recording. A high-pass filter is used to process the low-frequency baseline drift of breathing and heartbeat. The formula of wavelet transform can be expressed as:
[0112] S(t)=∑ j ∑ k c jk ψ jk (t) (2)
[0113] Where S(t) is the signal, c jk is the wavelet coefficient, ψ jk (t) is the wavelet basis function, j and k represent the decomposition level and position respectively.
[0114] S213, normalizing the physiological signals after baseline correction;
[0115] In a feasible implementation, the data needs to be normalized before being input into the neural network to ensure that all features are at the same scale, so as to avoid the slight pattern changes of some complex physiological signals (such as ECG) being ignored due to different scales, thereby speeding up the optimization process. In addition, normalization helps improve the generalization ability of the model on unseen data by reducing the sensitivity of the model to the scale of the input value. Z-score normalization can be expressed as:
[0116]
[0117] in, N is the total number of sampling points of the signal, X is the original data point, μ is the original data mean, σ is the standard deviation of the original data, and Z is the standardized data point.
[0118] S22, inputting the preprocessed multi-channel physiological signals of the operator into a bidirectional long short-term memory network based on a hybrid attention mechanism to obtain the emotion type and emotional arousal of the operator;
[0119] It should be noted that the present invention proposes a HABi-LSTM (Mixed Attention Mechanism with Bi-directional Long Short-Term Memory, a bidirectional long short-term memory network based on a hybrid attention mechanism) (Special note: Figure 3 The following is the structural diagram of the model, which shows the input, output, and structural composition of each key module of the model. The model can flexibly adjust the input signal dimension and form, the number and form of classification types of the output results, and the model depth according to the actual signal acquisition situation and specific needs. Therefore, there is no need to provide a network with fixed parameters and network layers. ) to achieve emotion type recognition and arousal evaluation. Compared with earlier methods, this structure enables the model to make full use of the spatiotemporal characteristics of the time series, reasonably allocate the contribution of each signal to the final task, and improve the accuracy and efficiency of identifying changes in emotional states at work. The model structure is shown in Figure 3 The core of the HABi-LSTM model consists of the Bi-LSTM layer, the internal attention layer, and the external attention layer. The following will introduce the role of each part in the model in detail.
[0120] Bi-LSTM processes input time series data, allowing the model to capture both forward and backward temporal dependencies, which is particularly suitable for physiological signal analysis involving temporal dynamics. t Processing, combined with the previous hidden state And the hidden state after to update the current status.
[0121]
[0122] Finally, the output of each time step is obtained by merging the hidden states of the forward and backward LSTMs.
[0123]
[0124] An internal attention mechanism is applied to each independent signal to determine the importance of each time point within each signal. A small neural network is used to learn the contribution of each time point to the final output, so that the model can focus on more informative time points, thereby improving learning efficiency. In this step, the output of the LSTM is first transformed through a linear layer and a nonlinear activation function, and then the attention weight of each time step is calculated, which is expressed as follows:
[0125] u t =tanh(W c h t +b c ) (8)
[0126]
[0127] Among them, W c is a weight matrix used to transform the hidden state to h t Attention space; T represents the transpose of the matrix; b c is the bias term; μ t is the transformation vector at time step t; α t is the normalized attention weight at time step t, indicating how much the model should pay attention to the output of this time step; tanh() represents the hyperbolic tangent activation function, which is used to introduce nonlinear transformation; u t' represents the transformation variable at time step t'; t' represents all possible time steps, which are used to calculate the normalized attention weights;
[0128] Then, use α t Get h t The weighted feature representation c is expressed as follows:
[0129] c=∑ t α t h t (10)
[0130] After obtaining the weighted feature representation of a single signal through the internal attention layer, the model applies a cross-signal attention mechanism to the output of different signals to evaluate the contribution of different types of signals to the recognition task. This step is different from the traditional random weighted fusion of multi-sensor data. Instead, it uses the model's own feedback to understand which signals provide more useful information in improving performance, and achieves targeted weighting of each signal:
[0131] β i =sigmoid(W s c i +b s ) (11)
[0132] s=Σ i β i c i (12)
[0133] Among them, W s and b s is a learning parameter; s is the weighted final integrated feature representation of each signal; sigmoid() represents an S-shaped activation function used to map the input to the range of 0 to 1; c i represents the feature representation of the i-th signal;
[0134] Finally, the output layer of the network uses two branches to simultaneously output the emotion type and emotional arousal;
[0135] The branch that outputs the emotion category is used to predict the probability distribution of the emotion category corresponding to the input data. The prediction process is as follows: Assuming that there are N emotion categories, s is mapped to a vector of length N through a fully connected layer, representing the score of each emotion category. Then, the Softmax activation function is used to convert the emotion category score into a probability distribution, so that the probability value of each category is in the range of [0,1], and the sum of the probabilities of all categories is 1;
[0136] emotion_logits=Linear(s) (13)
[0137]
[0138] Among them, P i represents the predicted probability of the i-th category of emotion.
[0139] The branch that outputs emotional arousal is used to predict the arousal value corresponding to the input data, which is usually a scalar value. The output s of the middle layer is mapped to a scalar value through a fully connected layer, representing the arousal value:
[0140] A=Linear(s) (15)
[0141] Where A represents the predicted arousal value.
[0142] S23, calculating the trust level of the current remote control device in the operator based on the emotion type and emotional arousal of the current operator;
[0143] Specifically, in this embodiment, the calculation process of the trust level is as follows:
[0144] S231, determining the trust score corresponding to the emotion type of the current operator and the trust score corresponding to the emotional arousal degree of the current operator;
[0145] S232, respectively setting the influence weights of emotion type and emotion arousal on the trust score;
[0146] In one feasible approach, it is assumed that the trust score ranges from 0 to 1, where 0 represents complete distrust and 1 represents complete trust. The weight of the influence of each emotion type and arousal degree on the trust score is determined based on experience or data analysis.
[0147] S233, calculating the trust level of the current remote control device to the operator, the formula is as follows:
[0148] P=W E ×S E +W A ×S A
[0149] Where P is the trust level of the current remote control device to the operator; W E is the influence weight of emotion type on trust score; S E is the trust score corresponding to the emotion type of the current controller (based on a predefined scoring table); W A is the influence weight of emotional arousal on trust score; S A It is the trust score corresponding to the emotional arousal of the current operator (normalized score based on the arousal value A).
[0150] S3, dynamically adjusting the control strategy of the remote control device based on the current trust level of the remote control device in the operator;
[0151] In a feasible implementation, the present invention proposes a user work state driven (User Work State-Driven, UWSD) control method. Traditional output control methods, such as proportional-integral-differential (ProportionalIntegral Derivative, PID) control, open-loop control and rule-based systems, etc., although they can provide stable performance under certain conditions, they are often unable to adapt to the dynamic changes of the environment when facing complex and changeable human-computer interaction environments and emergency response situations, especially unable to effectively deal with the operator's possible overreactions in emergency situations. The user work state driven (User Work State-Driven, UWSD) control method proposed in the present invention can automatically activate the corresponding safety protocol based on the user's work state perceived by the human-computer trust system, effectively adjust the operation strategy of the robot or remote control car, and reduce the risks and accidents caused by changes in the operator's work state. It can be manifested as when the system detects that the operator is in a state of panic due to an emergency, it temporarily takes over the control of the system and automatically adjusts the operation strategy of the robot or remote control vehicle to alleviate excessive operation reactions and ensure the safety of the operation process. Figure 4 The working mode of the UWSD control method is shown. For emergency events, the system has designed a special response mechanism. According to the real-time output of the user status perception module, the various operating parameters of the system can be adjusted. The working process is as follows:
[0152] S31, determine specific response measures:
[0153]
[0154] Among them, R represents the specific response measures taken according to the status score, and whether to trigger an emergency response is determined by setting the threshold H. Status 1 represents the triggering of the safety protocol, and status 0 represents normal working conditions.
[0155] S32, based on the state recognition result and the individual differences of the operators, a safety adjustment coefficient is calculated to adjust the operating parameters of the remote control device, such as the braking degree and steering strength. When calculating the safety adjustment coefficient, the weight can be dynamically allocated according to the current level of trust of the machine in the human and the individual differences of the operators:
[0156] W1=h1(P,S Individual Difference )=a1P+b1 (17)
[0157] W2=h2(P,S Individual Difference )=a2P+b2 (18)
[0158] W3=h3(P,S Individual Difference )=a3P+b3 (19)
[0159] W4=h4(P,S Individual Difference )=a4P+b4 (20)
[0160] Based on the assigned weights, the safety adjustment factor is calculated as follows:
[0161] S1=A(Emontion)×W1 (21)
[0162] S2=A(Emontion)×W2 (22)
[0163] S3=A(Emontion)×W3 (23)
[0164] S4=A(Emontion)×W4 (24)
[0165] Among them, h1, h2, h3 and h4 are weight functions determined according to the trust level of the current remote control device in the operator and the individual differences of the operators. a1, a2, a3 and a4 are adjustment factors in the weight function that can be set in the range of 0 to 1. b1, b2, b3 and b4 are used to adjust individual differences.
[0166] S Individual DifferenceThe set bias (for example, using a quick response task test to evaluate the operator's reaction time and determine their ability to adapt to speed changes. Evaluate the operator's sensitivity to speed changes by responding to sudden acceleration or deceleration in operation simulation or actual operation. According to the above test results, set the acceleration and deceleration rate bias for each operator. For example, quick responders may prefer faster acceleration and deceleration, while operators who are sensitive to speed changes require a smoother acceleration and deceleration curve. At the same time, collect feedback in actual operations, such as continuously optimizing the acceleration and deceleration rate settings through the operator's evaluation of emergency braking or rapid start.); W1, W2, W3 and W4 are weight coefficients. By adjusting the weights, different operating environments and user needs can be flexibly responded to; A (Emontion) represents the standardized value of emotional arousal.
[0167] S33, dynamically adjusting the operating parameters of the remote control system using the safety adjustment coefficient, specifically including:
[0168] S331, dynamically adjust the travel speed of the remote control system according to the safety adjustment coefficient to adapt to changes in the operating environment. In a feasible implementation, the control parameters after the operation are expressed by the following formula:
[0169] V'=V·(1-α·S1) (25)
[0170] Wherein, V' represents the travel speed of the device after adjustment; V represents the travel speed of the device before adjustment; α represents the weight coefficient, which can be set to a value of 0 to 1 according to the actual situation;
[0171] S332 dynamically adjusts the steering angle based on the safety adjustment factor to ensure that the remote control system can steer safely and accurately in complex or dangerous environments.
[0172] In a feasible implementation, the control parameters after the operation are expressed by the following formula:
[0173] θ'=θ·(1-β·S2) (26)
[0174] Among them, θ' represents the steering angle of the device after adjustment; θ represents the steering angle of the device before adjustment; β represents the weight coefficient, which can be set to a value of 0 to 1 according to the actual situation;
[0175] S333 dynamically adjusts the braking force based on the safety adjustment factor to ensure timely and effective deceleration or stopping when emergency braking is required.
[0176] In a feasible implementation, the control parameters after the operation are expressed by the following formula:
[0177] F'=F·(1-χ·S3) (27)
[0178] Among them, F' represents the braking force of the device after adjustment; F represents the braking force of the device before adjustment; χ represents the weight coefficient, which can be set to a value of 0 to 1 according to the actual situation;
[0179] S334 dynamically adjusts the acceleration and deceleration rates according to the safety adjustment factor, making the remote control system more stable during acceleration and deceleration and reducing sudden operations caused by emotional changes of the operator.
[0180] In a feasible implementation, the control parameters after the operation are expressed by the following formula:
[0181] a'=a·(1-δ·S4) (28)
[0182] Wherein, a' represents the acceleration and deceleration rate of the device after adjustment; a represents the acceleration and deceleration rate of the device before adjustment; δ represents the weight coefficient, which can be set to a value of 0 to 1 according to actual conditions.
[0183] In summary, this embodiment provides a method for dynamically adjusting the control strategy of a remote control device based on human-machine trust. The method can comprehensively and accurately collect multi-channel physiological signals of the remote control operator during work, which is conducive to the subsequent design and application of various physiological signal analysis and output control algorithms. Moreover, the method can monitor and mark the emotional state at the same time, and automatically adjust the system control strategy according to the current working state. It has the advantages of wearable, non-invasive, reliable information processing, and high degree of automation, which can significantly improve the operational stability and safety of the human-machine interaction system. In addition, the method comprehensively adopts the attention mechanism and long short-term memory network for emotional state recognition, and simultaneously uses four physiological signals of electrocardiogram, skin galvanic response, respiratory signal and pulse signal for learning, which has the advantages of high information utilization and accurate recognition results. It has high flexibility, the control output algorithm can be adjusted in real time according to the current state, and the corresponding control strategy can be given according to the different states of different operators. The experimental results show that the method significantly improves work efficiency and operation accuracy.
[0184] Second embodiment
[0185] This embodiment provides a remote control device control strategy dynamic adjustment system based on human-machine trust. The structure of the remote control device control strategy dynamic adjustment system based on human-machine trust is as follows: Figure 5 As shown, it includes the following modules:
[0186] A data acquisition module, used to acquire multi-channel physiological signals of the operator of the remote control device in the working state;
[0187] A remote control device trust level calculation module for the operator, used to calculate the current remote control device trust level for the operator based on the multi-channel physiological signals of the operator acquired by the data acquisition module;
[0188] The control strategy dynamic adjustment module is used to dynamically adjust the control strategy of the remote control device based on the current trust level of the remote control device to the operator calculated by the trust level calculation module of the remote control device to the operator.
[0189] Among them, it should be noted that the remote control device control strategy dynamic adjustment system based on human-machine trust of this embodiment corresponds to the remote control device control strategy dynamic adjustment method based on human-machine trust of the above-mentioned first embodiment; wherein, the functions implemented by each functional module in the remote control device control strategy dynamic adjustment system based on human-machine trust of this embodiment correspond one by one to each process step in the remote control device control strategy dynamic adjustment method based on human-machine trust of the above-mentioned first embodiment; therefore, they will not be repeated here.
[0190] Third embodiment
[0191] This embodiment provides an electronic device, such as Figure 6 As shown, the electronic device includes: a processor and a memory; wherein the processor and the memory can be connected via a communication bus; the memory stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method of the first embodiment. In addition, the electronic device may also include a transceiver, the processor and the transceiver can be connected via a communication bus, and the transceiver is used to communicate with other devices.
[0192] Next, combine Figure 6 The following is a detailed introduction to the various components of the electronic device:
[0193] Among them, the processor is the control center of the electronic device, and the electronic device may include multiple processors, each of which may be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here may be a processor or a general term for multiple processing elements. For example, the processor is one or more central processing units (CPUs), or other general-purpose processors, application specific integrated circuits (ASICs), or one or more integrated circuits configured to implement an embodiment of the present invention, such as one or more microprocessors (digital signal processors, DSPs), or one or more field programmable gate arrays (field programmable gate arrays, FPGAs), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor may execute various functions of the electronic device by running or executing software programs stored in the memory and calling data stored in the memory.
[0194] In a specific implementation, as an embodiment, the processor may include one or more CPUs, such as Figure 6 The CPU0 and CPU1 shown in the figure are, of course, only exemplary.
[0195] The memory is used to store the software program for executing the solution of the present invention, and the execution is controlled by the processor. The specific implementation method can refer to the above method embodiment, which will not be repeated here.
[0196] Optionally, the memory may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may be integrated with the processor or exist independently and accessed through the interface circuit ( Figure 6 (not shown) is coupled to the processor, which is not specifically limited in this embodiment of the present invention.
[0197] The transceiver may include a receiver and a transmitter ( Figure 6 The receiver is used to implement the receiving function, and the transmitter is used to implement the sending function. The transceiver can be integrated with the processor or exist independently and communicate with the electronic device through the interface circuit ( Figure 6 (not shown) is coupled to the processor, which is not specifically limited in this embodiment of the present invention.
[0198] In addition, it should be noted that Figure 6 The structure of the electronic device shown in the figure does not constitute a limitation on the device, and the actual device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. In addition, the technical effects achieved by the electronic device when executing the method of the first embodiment above can refer to the technical effects described in the first embodiment above, so they are not repeated here.
[0199] Fourth embodiment
[0200] This embodiment provides a computer-readable storage medium, which stores at least one instruction, and the instruction is loaded and executed by a processor to implement the method of the first embodiment. The computer-readable storage medium may be a ROM, a random access memory, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc. The instructions stored therein may be loaded by a processor in a terminal to execute the method.
[0201] In addition, it should be noted that the present invention can be provided as a method, an apparatus or a computer program product. Therefore, the embodiment of the present invention can be in the form of a full or partial hardware embodiment, a full or partial software embodiment or an embodiment combining software and hardware. Moreover, when implemented using software, the embodiment of the present invention can be in the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program codes. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center containing one or more available media sets. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium. The semiconductor medium may be a solid state hard disk.
[0202] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0203] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable terminal device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0204] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of more restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements. In addition, the term "and / or" is only an association relationship describing the associated objects, indicating that there can be three relationships, for example, A and / or B, which can represent: A exists alone, A and B exist at the same time, and B exists alone, wherein A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding. "At least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can be represented by: a, b, c, ab, ac, bc or abc, where a, b, c can be single or plural.
[0205] In addition, it can be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0206] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0207] In several embodiments provided by the present invention, it should be understood that the disclosed equipment, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of functional modules / units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point, the coupling or direct coupling or communication connection between each other shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.
[0208] If the method is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0209] Finally, it should be noted that the above is only a preferred embodiment of the present invention. It should be pointed out that although the preferred embodiment of the present invention has been described, for ordinary technicians in this technical field, once the basic creative concept of the present invention is known, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the protection scope of the present invention. Therefore, the attached claims are intended to be interpreted as including the preferred embodiment and all changes and modifications that fall within the scope of the embodiments of the present invention.
Claims
1. A method for dynamically adjusting the control strategy of remote control equipment based on human-machine trust, It is characterized in that The method for dynamically adjusting the control strategy of a remote control device based on human-machine trust includes: Acquire multi-channel physiological signals of operators of remote control equipment in working state; Based on the multi-channel physiological signals of the operator, calculate the trust level of the current remote control device to the operator; Based on the current trust level of the remote control device in the operator, the control strategy of the remote control device is dynamically adjusted.
2. The method for dynamically adjusting the control strategy of a remote control device based on human-machine trust as claimed in claim 1, It is characterized in that The multi-channel physiological signal of the operator includes: an electrocardiogram signal of the operator, an electrical skin activity signal of the operator, a breathing signal of the operator and a pulse signal of the operator.
3. The method for dynamically adjusting the control strategy of a remote control device based on human-machine trust as claimed in claim 2, It is characterized in that The calculating the trust level of the current remote control device to the operator based on the multi-channel physiological signal of the operator includes: Preprocessing the acquired multi-channel physiological signals of the operator; wherein the preprocessing includes: signal denoising, baseline correction and standardization; The pre-processed multi-channel physiological signals of the operator are input into a bidirectional long short-term memory network based on a hybrid attention mechanism to obtain the operator's emotion type and emotional arousal; Based on the emotion type and emotional arousal of the current operator, the trust level of the current remote control device in the operator is calculated.
4. The method for dynamically adjusting the control strategy of a remote control device based on human-machine trust as claimed in claim 3, It is characterized in that The preprocessing of the acquired multi-channel physiological signals of the operator includes: The electrocardiogram signal is filtered by a bandpass filter, the baseline drift of the filtered electrocardiogram signal is removed by wavelet transform, the baseline correction of the electrocardiogram signal is realized, and the zero mean normalization algorithm is used to standardize the electrocardiogram signal after the baseline correction; The skin electrical activity signal is filtered by a low-pass filter, the baseline drift of the filtered skin electrical activity signal is removed by wavelet transform, the baseline correction of the skin electrical activity signal is achieved, and the baseline-corrected skin electrical activity signal is standardized by a zero-mean normalization algorithm; A low-pass filter is used to filter the respiratory signal, a high-pass filter is used to process the low-frequency baseline drift of the filtered respiratory signal, and a baseline correction of the respiratory signal is achieved. A zero-mean normalization algorithm is used to standardize the respiratory signal after baseline correction; A low-pass filter is used to filter the pulse signal, and a high-pass filter is used to process the low-frequency baseline drift of the filtered pulse signal to achieve baseline correction of the pulse signal. A zero-mean normalization algorithm is used to standardize the pulse signal after baseline correction.
5. The method for dynamically adjusting the control strategy of a remote control device based on human-machine trust as claimed in claim 3, It is characterized in that The bidirectional long short-term memory network based on the hybrid attention mechanism includes: an input layer, a Bi-LSTM layer, an internal attention layer, an external attention layer and an output layer; The process of processing input data by the bidirectional long short-term memory network based on the hybrid attention mechanism includes: The Bi-LSTM layer is used to process the time series data inputted by the input layer at each time step to obtain the output of each time step; and the output of each time step is respectively transmitted to the internal attention layer; The internal attention layer is used to determine the importance of each time point within each independent signal; the internal attention layer processes the data by first transforming the output of the Bi-LSTM layer through a linear layer and a nonlinear activation function, and then calculating the attention weight of each time step, which is expressed as follows: u t =tanh(W c h t +b c ) Among them, W c is a weight matrix used to transform the hidden state to the output h of the Bi-LSTM layer at the current time step t t The attention space of T represents the transpose of the matrix; b c is the bias term; tanh() represents the hyperbolic tangent activation function, which is used to introduce nonlinear transformation; u t' represents the transformation vector at time step t'; t' represents all possible time steps, which are used to calculate the normalized attention weights; Then, use α t Get h t The weighted feature representation c is expressed as follows: c=∑ t a t h t After obtaining the weighted feature representation of a single signal through the internal attention layer, the external attention layer is used to evaluate the contribution of different types of signals to the recognition task; the expression is as follows: β i =sigmoid(W s c i +b s ) s=∑ i b i c i Among them, W s and b s is a learning parameter; sigmoid() represents an S-shaped activation function used to map the input to the range of 0 to 1; c i represents the feature representation of the i-th signal; Finally, the output layer of the network uses two branches to simultaneously output the emotion type and emotional arousal; The branch that outputs the emotion category is used to predict the probability distribution of the emotion category corresponding to the input data. The prediction process is as follows: Assuming that there are N emotion categories, s is mapped to a vector of length N through a fully connected layer, representing the score of each emotion category. Then, the Softmax activation function is used to convert the emotion category score into a probability distribution, so that the probability value of each category is in the range of [0,1], and the sum of the probabilities of all categories is 1; The branch that outputs the emotional arousal is used to predict the emotional arousal corresponding to the input data. The prediction process is: s is mapped to a scalar value representing the emotional arousal through a fully connected layer.
6. The method for dynamically adjusting the control strategy of a remote control device based on human-machine trust as claimed in claim 3, It is characterized in that The step of calculating the trust level of the current remote control device in the operator based on the emotion type and emotional arousal of the current operator includes: The trust score corresponding to the emotion type of the current operator and the trust score corresponding to the emotional arousal degree of the current operator are set based on a preset scoring table; wherein the scoring table includes trust scores corresponding to different types of emotions and trust scores corresponding to different emotional arousal degrees; Set the weights of emotion type and emotion arousal on trust score respectively; Calculate the trust level of the current remote control device to the operator, the formula is as follows: P=W E ×S E +W A ×S A Where P is the trust level of the current remote control device to the operator; W E is the influence weight of emotion type on trust score; S E is the trust score corresponding to the current controller’s emotion type; W A is the influence weight of emotional arousal on trust score; S A It is the trust score corresponding to the emotional arousal of the current controller.
7. The method for dynamically adjusting the control strategy of a remote control device based on human-machine trust as claimed in claim 1, It is characterized in that The dynamically adjusting the control strategy of the remote control device based on the current trust level of the remote control device to the operator includes: The current trust level of the remote control device in the operator is compared with the preset threshold; if the current trust level of the remote control device in the operator is not less than the preset threshold, the operator continues to control the device; otherwise, based on the current trust level of the remote control device in the operator and combined with the individual differences of the operators, a safety adjustment coefficient is calculated, and the operating status of the device is adjusted based on the safety adjustment coefficient.
8. The method for dynamically adjusting the control strategy of a remote control device based on human-machine trust as claimed in claim 7, It is characterized in that The operating status of the device includes: the travel speed of the device, the steering angle of the device, the braking force of the device, and the acceleration and deceleration rates of the device.
9. The method for dynamically adjusting the control strategy of a remote control device based on human-machine trust as claimed in claim 8, It is characterized in that According to the current trust level of the remote control device in the operator and the individual differences of the operators, the safety adjustment factor is calculated, and the operating status of the device is adjusted based on the safety adjustment factor, including: According to the current trust level of the remote control device in the operator, combined with the individual differences of the operators, weights are dynamically allocated. The expressions of each weight are as follows: W 1 =a 1 P+b 1 W 2 =a 2 P+b 2 W 3 =a 3 P+b 3 W 4 =a 4 P+b 4 Based on the assigned weights, the safety adjustment factor is calculated as follows: S 1 =A(Emontion)×W 1 S 2 =A(Emontion)×W 2 S 3 =A(Emontion)×W 3 S 4 =A(Emontion)×W 4 Where P is the trust level of the current remote control device to the operator; a 1 , a 2 , a 3 and a 4 is the adjustment factor; b 1 , b 2 , b 3 and b 4 is the bias; A(Emontion) represents the standardized value of emotional arousal. Based on S 1 Adjust the travel speed of the equipment; based on S 2 Adjust the steering angle of the device; based on S 3 Adjust the braking force of the device; based on S 4 Adjusts the rate at which the device accelerates and decelerates.
10. The method for dynamically adjusting the control strategy of a remote control device based on human-machine trust according to claim 9, It is characterized in that The adjustment formula for the equipment's travel speed is: V'=V·(1-α·S 1 ) Wherein, V' represents the travel speed of the device after adjustment; V represents the travel speed of the device before adjustment; α represents the weight coefficient; The adjustment formula for the steering angle of the device is: θ'=θ·(1-β·S 2 ) Wherein, θ' represents the steering angle of the device after adjustment; θ represents the steering angle of the device before adjustment; β represents the weight coefficient; The adjustment formula for the braking force of the equipment is: F'=F·(1-x·S 3 ) Wherein, F' represents the braking force of the device after adjustment; F represents the braking force of the device before adjustment; χ represents the weight coefficient; The adjustment formula for the acceleration and deceleration rate of the equipment is: a'=a·(1-δ·S 4 ) Wherein, a' represents the acceleration and deceleration rate of the device after adjustment; a represents the acceleration and deceleration rate of the device before adjustment; δ represents the weight coefficient.
