Shared control method and related device for man-machine co-driving
By acquiring the driver's brainwave signals and driving data, and using bargaining game theory to allocate driving permissions, the problems of low collaborative efficiency and uneven interaction in the human-machine co-driving mode are solved, thereby improving safety, robustness and flexibility.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- WUHAN UNIV OF TECH
- Filing Date
- 2026-02-12
- Publication Date
- 2026-06-05
Smart Images
Figure CN122143947A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of intelligent driving technology, and in particular to a human-machine co-driving and shared control method and related equipment. Background Technology
[0002] With continuous breakthroughs in core technologies such as artificial intelligence, multi-source information perception, and multi-agent collaborative control, automotive intelligence has become a major development direction for modern transportation systems. In the transition to fully autonomous driving (Level 4-5), shared driving is a technological architecture that organically integrates the cognitive advantages of human drivers with the precise control capabilities of autonomous driving systems. Due to differences between human drivers and machine systems in environmental perception, decision-making logic, and action execution, inconsistencies in commands regarding the same control objective (such as steering wheel angle) inevitably lead to human-machine conflict. Related technologies often struggle to strike a good balance between the machine's obstacle avoidance safety requirements and its compatibility with driver operating habits, resulting in low collaborative efficiency and an unnatural and unsmooth human-machine interaction process. Furthermore, shared driving suffers from poor safety, robustness, and flexibility.
[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention
[0004] The main objective of this application is to propose a human-machine co-driving shared control method and related equipment, which can effectively alleviate the problems of low collaborative efficiency and insufficient smoothness of human-machine interaction in human-machine co-driving mode, and effectively improve the safety, robustness and flexibility of human-machine co-driving.
[0005] To achieve the above objectives, one aspect of this application proposes a human-machine co-driving and shared control method, the method comprising: The driver's brainwave signal data is acquired through a preset brain-computer interface, and then the brainwave signal data is analyzed by a human-machine conflict monitoring unit to determine the conflict index data. Preset driving data is acquired, and then the vehicle safety status assessment unit performs a vehicle collision risk assessment on the preset driving data to obtain safety benefit data. Based on the conflict index data and the safety benefit data, driving authority allocation analysis is performed through the shared control unit to obtain the desired control authority allocation parameters; wherein, the shared control unit is constructed based on bargaining game theory; Based on the desired control permission allocation parameters and the preset reference path, path planning is performed to generate a target driving path; wherein, the preset reference path includes a driver reference path and an autonomous driving reference path.
[0006] In some embodiments, the step of acquiring the driver's electroencephalogram (EEG) signal data through a preset brain-computer interface, and then analyzing the EEG signal data through a human-machine conflict monitoring unit to determine conflict index data, includes: Error-related potential data is acquired through the preset brain-computer interface, and conflict feature data is then extracted from the error-related potential data. The conflict feature data is identified by a preset machine learning model, and then the conflict quantization calculation is performed based on the identified conflict frequency band signals to obtain the conflict index data.
[0007] In some embodiments, extracting conflict feature data from the error-related potential data includes: The error-related potential data is filtered by a preset bandpass filter to obtain the first EEG signal; The first EEG signal was processed by an independent component analysis algorithm to remove artifacts, resulting in the second EEG signal. A spatial filter is constructed using a common spatial pattern algorithm, and then the spatial filter is used to extract spatial features from the second EEG signal to obtain the conflict feature data.
[0008] In some embodiments, before performing the acquisition of preset driving data and then using a vehicle safety status assessment unit to perform a vehicle collision risk assessment on the preset driving data to obtain safety benefit data, the method further includes: Construct a vehicle dynamics model; wherein the vehicle dynamics model includes air resistance and environmental disturbance terms; A preset trajectory tracker is constructed based on the linear quadratic regulator algorithm.
[0009] In some embodiments, the step of acquiring preset driving data and then using a vehicle safety status assessment unit to assess the vehicle collision risk of the preset driving data to obtain safety benefit data includes: Acquire the preset driving data; wherein the preset driving data includes vehicle status information and environmental obstacle information; Based on the vehicle status information and the environmental obstacle information, a collision risk analysis is performed using the vehicle dynamics model and the preset trajectory tracker to obtain preset risk assessment data; wherein, the preset risk assessment data includes longitudinal collision risk data, lateral collision time data, and forward time data; The preset risk assessment data is transformed into a comprehensive risk assessment index through a nonlinear correction function, and then the safety benefit data is determined based on the comprehensive risk assessment index.
[0010] In some embodiments, the step of performing driving permission allocation analysis through a shared control unit based on the conflict index data and the safety benefit data to obtain desired control permission allocation parameters includes: Construct a preset multidimensional sub-benefit function; wherein, the preset multidimensional sub-benefit function includes a conflict sub-benefit function, a safety sub-benefit function, a participation sub-benefit function, and a comfort sub-benefit function; Construct a total benefit function based on the conflict sub-benefit function, the safety sub-benefit function, the participation sub-benefit function, and the comfort sub-benefit function; Based on the conflict index data and the security benefit data, the total benefit function is solved using Nash equilibrium to determine the expected control authority allocation parameters.
[0011] To achieve the above objectives, another aspect of this application provides a human-machine co-driving and shared control device, the device comprising: The first module is used to acquire the driver's brainwave signal data through a preset brain-computer interface, and then analyze the brainwave signal data through a human-machine conflict monitoring unit to determine the conflict index data. The second module is used to acquire preset driving data, and then use the vehicle safety status assessment unit to assess the vehicle collision risk of the preset driving data to obtain safety benefit data. The third module is used to perform driving permission allocation analysis through the shared control unit based on the conflict index data and the safety benefit data to obtain the desired control permission allocation parameters; wherein, the shared control unit is constructed based on bargaining game theory; The fourth module is used to perform path planning based on the desired control permission allocation parameters and preset reference paths to generate a target driving path; wherein the preset reference paths include a driver reference path and an autonomous driving reference path.
[0012] To achieve the above objectives, another aspect of this application provides an electronic device, the electronic device comprising: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor performs the method described above.
[0013] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.
[0014] To achieve the above objectives, another aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method. The embodiments of this application include at least the following beneficial effects: This application provides a human-machine co-driving and shared control method, device, electronic device, storage medium, and program product. This solution acquires the driver's electroencephalogram (EEG) signal data through a preset brain-computer interface, and then analyzes the EEG signal data through a human-machine conflict monitoring unit to determine conflict index data. Simultaneously, the embodiments of this invention acquire preset driving data, and use a vehicle safety status assessment unit to assess the vehicle collision risk of the preset driving data to obtain safety benefit data. Then, based on the conflict index data and safety benefit data, a shared control unit constructed based on bargaining game theory is used to analyze driving permission allocation to obtain desired permission allocation parameters. Based on the desired control permission allocation parameters and a preset reference path, including a driver reference path and an autonomous driving reference path, path planning is performed to generate a target driving path, thereby realizing human-machine co-driving and shared control. It is easy to understand that the embodiments of the present invention can effectively alleviate the problem of delayed conflict perception by collecting EEG signals to determine conflict index data, realize real-time quantitative feedback of conflict, and effectively alleviate the problem of low collaborative efficiency and insufficient smoothness of human-machine interaction in human-machine co-driving mode by combining conflict index data and safety benefit data through a shared control unit constructed based on bargaining game theory. Furthermore, it can effectively improve the safety, robustness, and flexibility of human-machine co-driving. Attached Figure Description
[0015] Figure 1 This is a flowchart of the human-machine co-driving and shared control method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the vehicle route provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of vehicle driving authority provided in an embodiment of the present invention; Figure 4 This is a box plot of vehicle position error provided in an embodiment of the present invention; Figure 5 This is a typical single-stimulation electrode waveform diagram of error-related potential (ErrP) provided in an embodiment of the present invention; Figure 6 This is a typical error-related potential (ErrP) continuous stimulation electrode waveform diagram provided in the embodiments of the present invention; Figure 7 This is a single-stimulation EEG topography map provided in an embodiment of the present invention; Figure 8 This is a continuous stimulation electroencephalogram (EEG) topography provided in an embodiment of the present invention; Figure 9This is a schematic diagram of the game logic architecture for human-machine co-driving permission allocation provided in an embodiment of the present invention; Figure 10 This is a schematic diagram of the overall architecture of human-machine co-driving based on ErrP monitoring and game theory provided in an embodiment of the present invention; Figure 11 This is a schematic diagram of the structure of the human-machine co-driving and shared control device provided in an embodiment of the present invention; Figure 12 This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation
[0016] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of apparatuses and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.
[0017] It is understood that the terms “first,” “second,” etc., used in this application may be used herein to describe various concepts, but unless otherwise stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the words “if,” “when,” or “in response to a determination” as used herein may be interpreted as “when…” or “when…” or “in response to a determination.”
[0018] As used in this application, the terms "at least one", "multiple", "each", "any", etc., "at least one" includes one, two or more, "multiple" includes two or more, "each" refers to each of the corresponding multiples, and "any" refers to any one of the multiples.
[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.
[0020] Before providing a detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained first. The nouns and terms involved in the embodiments of this application are subject to the following interpretations.
[0021] Electroencephalogram (EEG) refers to the weak electrical signals generated by the activity of neurons in the brain. The potential changes recorded through the scalp or inside the brain can reflect the real-time functional state of the brain.
[0022] Linear Quadratic Regulator (LQR): This is an optimal control method based on a linear system model and quadratic performance index. For linear time-invariant systems, it constructs a state feedback controller so that the weighted quadratic index, which combines the state error and control input energy, reaches its minimum value during the process from the initial state to the equilibrium state.
[0023] With continuous breakthroughs in core technologies such as artificial intelligence, multi-source information perception, and multi-agent collaborative control, automotive intelligence has become a major development direction for modern transportation systems. In the transition to fully autonomous driving (Level 4-5), shared driving is a technological architecture that organically integrates the cognitive advantages of human drivers with the precise control capabilities of autonomous driving systems. Due to differences between human drivers and machine systems in environmental perception, decision-making logic, and action execution, inconsistencies in commands regarding the same control objective (such as steering wheel angle) inevitably lead to human-machine conflict. Related technologies often struggle to strike a good balance between the machine's obstacle avoidance safety requirements and its compatibility with driver operating habits, resulting in low collaborative efficiency and an unnatural and unsmooth human-machine interaction process. Furthermore, shared driving suffers from poor safety, robustness, and flexibility.
[0024] In view of this, this application provides a human-machine co-driving shared control method, device, electronic device, storage medium, and program product. This solution acquires the driver's electroencephalogram (EEG) signal data through a preset brain-computer interface, and then analyzes the EEG signal data through a human-machine conflict monitoring unit to determine conflict index data. Simultaneously, this embodiment acquires preset driving data, and uses a vehicle safety status assessment unit to assess vehicle collision risk, obtaining safety benefit data. Then, based on the conflict index data and safety benefit data, a shared control unit constructed based on bargaining game theory is used to analyze driving permission allocation, obtaining desired permission allocation parameters. Based on these parameters and a preset reference path (including a driver reference path and an autonomous driving reference path), path planning is performed to generate a target driving path, achieving human-machine co-driving shared control. This effectively alleviates the problems of low collaborative efficiency and insufficient smoothness of human-machine interaction in human-machine co-driving mode, and effectively improves the safety, robustness, and flexibility of human-machine co-driving.
[0025] The human-machine co-driving and sharing control method provided in this application relates to the field of intelligent driving technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or in-vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the human-machine co-driving and sharing control method, but is not limited to the above forms.
[0026] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.
[0027] Figure 1 This is an optional flowchart of the human-machine co-driving and shared control method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S110 to S140.
[0028] Step S110: Obtain the driver's brainwave signal data through a preset brain-computer interface, and then analyze the brainwave signal data through a human-machine conflict monitoring unit to determine the conflict index data.
[0029] Step S120: Obtain preset driving data, and then use the vehicle safety status assessment unit to assess the vehicle collision risk of the preset driving data to obtain safety benefit data.
[0030] Step S130: Based on the conflict index data and safety benefit data, analyze the allocation of driving rights through the shared control unit to obtain the desired control rights allocation parameters. The shared control unit is constructed based on bargaining game theory.
[0031] Step S140: Based on the desired control permission allocation parameters and the preset reference path, perform path planning to generate the target driving path. The preset reference path includes a driver reference path and an autonomous driving reference path.
[0032] In this specific embodiment, the present invention first acquires the driver's electroencephalogram (EEG) signal data through a preset brain-computer interface, and then analyzes the EEG signal data through a human-machine conflict monitoring unit to determine conflict index data. Specifically, in the shared driving mode, the core cause of human-machine conflict lies in the deviation between the machine's decision-making instructions and the driver's psychological expectations. Therefore, the present invention uses brain-computer interface technology to monitor the characteristic bioelectrical signals emitted by the driver when they perceive the system's erroneous decision, i.e., EEG signal data. Accordingly, the human-machine conflict monitoring unit in this embodiment is constructed using an optimized machine learning algorithm. Then, the present invention uses the human-machine conflict monitoring unit to extract features and classify the EEG signal data to determine the corresponding conflict index data. Here, the conflict index data refers to the quantified human-machine conflict degree data. Simultaneously, the present invention acquires preset driving data, and then uses a vehicle safety status assessment unit to assess the vehicle collision risk based on the preset driving data to obtain safety benefit data. Specifically, the preset driving data in this embodiment refers to the vehicle's state data during driving, such as vehicle state and environmental state. Accordingly, the vehicle safety status assessment unit in this embodiment is constructed based on the vehicle's dynamics model and control model. This invention analyzes preset driving data through a vehicle safety status assessment unit to determine whether the vehicle has a collision risk and obtain corresponding safety benefit data. The safety benefit data reflects the degree of danger in the current environment. Further, this invention analyzes driving permission allocation through a shared control unit based on conflict index data and safety benefit data to obtain desired control permission allocation parameters. Then, based on the desired control permission allocation parameters and a preset reference path, path planning is performed to obtain the target driving path. Specifically, in this invention, the shared control unit is constructed based on bargaining game theory, transforming the allocation of human-machine co-driving permissions into a game problem of dynamic interaction between two intelligent agents. Accordingly, after analyzing driving permissions by combining conflict index data and safety benefit data to obtain the optimal driving permission allocation ratio (i.e., desired control permission allocation parameters), this invention performs path planning based on the preset reference paths, including the driver reference path and the autonomous driving reference path, according to these desired control permission allocation parameters, i.e., allocating permissions to the driver reference path and the autonomous driving reference path according to the desired control permission allocation parameters to generate the target driving path. The vehicle path, vehicle driving permissions, and vehicle position error box plot are shown below. Figure 2 , Figure 3 as well as Figure 4 As shown. Furthermore, to avoid drastic changes in control authority due to sensor fluctuations or sudden conflicts, this embodiment of the invention introduces a first-order low-pass filter to filter the solved desired control authority allocation parameters, such as the authority allocation coefficients. To ensure a smooth transition between permissions, a smooth processing method is applied. In this embodiment of the invention, permissions are allocated not only at the execution level but also through anthropomorphic replanning at the planning level. Accordingly, the execution path in this embodiment of the invention... That is, the target driving route, which is the driver's reference route. With autonomous driving reference path Based on the game's power value The weighted composition is shown in the following formula:
[0033] In this embodiment of the invention, the mechanism ensures that the generated trajectory not only meets the hard safety constraints of obstacle avoidance, but also conforms to the driver's operating habits to the greatest extent possible within the margin of safety.
[0034] In some embodiments of the present invention, the driver's electroencephalogram (EEG) signal data is acquired through a preset brain-computer interface, and then the EEG signal data is analyzed by a human-machine conflict monitoring unit to determine conflict index data, including but not limited to the following steps: Error-related potential data is acquired through a pre-defined brain-computer interface, and conflict feature data is then extracted from the error-related potential data.
[0035] The conflict feature data is identified by a pre-set machine learning model, and then the conflict quantification is calculated based on the identified conflict frequency band signals to obtain conflict index data.
[0036] In this specific embodiment, the present invention first acquires error-related potential (ErrP) data through a preset brain-computer interface to extract conflict feature data. Then, a preset machine learning model is used to identify the conflict feature data, and conflict quantification is performed based on the identified conflict frequency band signals to obtain conflict index data. Specifically, error-related potential (ErrP) signals mainly originate from the anterior cingulate cortex (ACC) region of the brain and have significant temporal characteristics. In human-machine collaborative driving scenarios, when the autonomous driving system performs steering operations and the decision does not meet the driver's expectations, it induces neurophysiological responses of error-related negative waves (ERN) and error-related positive waves (Pe) in the driver's cerebral cortex. The error-related negative wave typically appears first about 90ms after the error feedback is presented, exhibiting a characteristic negative deflection wave; the error-related positive wave is subsequently generated, reaching its peak within about 200-500ms after the error feedback is presented, and is related to the individual's error awareness and cognitive adjustment. The electrode waveforms of typical single and continuous ErrP stimulation are shown below. Figure 5 and Figure 6 As shown, the EEG topography maps for single and continuous stimuli are respectively as follows: Figure 7 and Figure 8As shown. Accordingly, this embodiment of the invention extracts effective conflict feature data from the original EEG data, i.e., error-related potential data, and decodes and classifies the conflict feature data through a preset machine learning model. For example, this embodiment of the invention uses the Support Vector Machine (SVM) algorithm to solve for the optimal separating hyperplane through the principle of structured risk minimization. Experimental verification shows that, under continuous error stimulus scenarios, the average classification accuracy of the Radial Basis Function Support Vector Machine (RBF-SVM) can reach 71.41%, effectively distinguishing between the two states of "cooperation and consistency" and "human-machine conflict". Finally, in order to quantify the conflict index and provide numerical support for subsequent game theory models, this embodiment of the invention introduces Event Related Spectrum Perturbations (ERSPs) to quantify the degree of conflict. Specifically, this embodiment of the invention first calculates the power enhancement of the conflict frequency band (4-7Hz and above 30Hz) relative to the baseline level and sets a conflict threshold. ( , The baseline mean and standard deviation are used to accumulate significant energy points and then normalize them, as shown in the following formula:
[0037] Where, in the formula It is the normalized value. It is a value obtained by accumulating the "significant energy points" that are in the "conflict frequency band" and exceed the critical value; Within the statistical period sample set The minimum value, Within the statistical period sample set The maximum value. Correspondingly, The larger the value, the more severe the deviation of the driver's subjective perception of intent, and this is used as the underlying data for constructing the conflict sub-benefits.
[0038] In some embodiments of the present invention, conflict feature data is extracted from error-related potential data, including but not limited to the following steps: The error-related potential data is filtered by a preset bandpass filter to obtain the first EEG signal.
[0039] The first EEG signal was processed by an independent component analysis algorithm to remove artifacts, resulting in the second EEG signal.
[0040] A spatial filter is constructed using a common spatial pattern algorithm, and then spatial features are extracted from the second EEG signal using the spatial filter to obtain conflict feature data.
[0041] In this specific embodiment, the present invention first filters the error-related potential (ERP) data using a preset bandpass filter to obtain a first EEG signal. Specifically, the original EEP data in this embodiment contains high noise and requires data preprocessing to extract conflict feature data. Specifically, the present invention first filters the original EEP data using a preset bandpass filter, such as a 1-9Hz bandpass filter, retaining the mu and theta rhythm signals closely related to ErrP, thus obtaining the first EEG signal. Next, the present invention uses an independent component analysis (ICA) algorithm to remove artifacts from the first EEG signal, obtaining a second EEG signal. Then, a spatial filter constructed using a common spatial pattern (CSP) algorithm is used to extract spatial features from the second EEG signal, obtaining conflict feature data. Specifically, the present invention uses independent component analysis (ICA) to identify and remove non-neuronal artifacts such as eye movements and electromyography (EMG), and uses a common spatial pattern (CSP) algorithm to construct a spatial filter. Specifically, by calculating the covariance matrix of correct and incorrect trials, a matrix orthogonal whitening transformation is performed to construct a spatial projection matrix with maximum class discrimination. This allows us to extract discriminative features that reflect the divergence between human and machine intentions, namely conflict feature data.
[0042] In some embodiments of the present invention, before acquiring preset driving data and then performing a vehicle collision risk assessment on the preset driving data through a vehicle safety status assessment unit to obtain safety benefit data, the human-machine co-driving and sharing control method provided in the embodiments of the present invention further includes, but is not limited to, the following steps: Construct a vehicle dynamics model. The vehicle dynamics model includes air resistance and environmental disturbance terms.
[0043] A preset trajectory tracker is constructed based on the linear quadratic regulator algorithm.
[0044] In this specific embodiment, the present invention first constructs a vehicle dynamics model including air resistance and environmental disturbance terms, and then constructs a preset trajectory tracker based on a linear quadratic regulator algorithm. Specifically, to provide the robot driver with a reasonable decision-making basis, the present invention establishes a vehicle physical model including wind resistance terms and multi-degree-of-freedom coupling. For example, the present invention constructs vehicle dynamics equations including lateral, longitudinal, and yaw motions, as shown in the following equation:
[0045] Where, in the formula It is the lateral displacement of the vehicle. , These are lateral velocity and lateral acceleration; It is the vehicle's yaw angle. , These are the yaw rate and yaw acceleration; It is the total mass of the vehicle; It is the longitudinal speed of the vehicle; It is the yaw moment of inertia of the vehicle about the Z-axis; , It is the distance from the vehicle's center of gravity to the front and rear axles; , It refers to the lateral stiffness of the front and rear wheels; It's the front wheel steering angle.
[0046] In addition, the embodiments of the present invention incorporate tire lateral stiffness. Establish state-space equations to describe lateral velocities. With yaw rate The evolution law is studied, and the robustness of the control algorithm in high-speed scenarios is ensured by incorporating air resistance and environmental disturbance terms into the model.
[0047] Accordingly, this embodiment of the invention constructs a discrete linear quadratic lateral trajectory tracking controller, i.e., a preset trajectory tracker, based on the LQR (Linear Quadratic Regulator) algorithm, as shown in the following equation:
[0048] Where, in the formula yes The state vector at any given time; This represents the predicted state value at the next moment; It is the control input at any given time; It is the state matrix of a discrete system, which describes the evolution of the system's state over time when there is no external control. It is a discrete control matrix that describes the control input. Impact on system state. Among them... , ( (The identity matrix).
[0049] Meanwhile, the embodiments of the present invention define an error state vector. This includes positional and heading deviations. Additionally, a quadratic performance index function is constructed. By solving discrete The equation yields the optimal feedback gain matrix. The final generated machine reference front wheel steering angle is... As one of the inputs for game decision-making.
[0050] In some embodiments of the present invention, preset driving data is obtained, and then a vehicle collision risk assessment is performed on the preset driving data by a vehicle safety status assessment unit to obtain safety benefit data, including but not limited to the following steps: Acquire preset driving data. This preset driving data includes vehicle status information and environmental obstacle information.
[0051] Based on vehicle status information and environmental obstacle information, a collision risk analysis is performed using a vehicle dynamics model and a preset trajectory tracker to obtain preset risk assessment data. This preset risk assessment data includes longitudinal collision risk data, lateral collision time data, and forward collision time data.
[0052] The preset risk assessment data is transformed into a comprehensive risk assessment index through a nonlinear correction function, and then the safety benefit data is determined based on the comprehensive risk assessment index.
[0053] In this specific embodiment, the present invention acquires preset driving data and performs collision risk analysis based on the preset driving data, combined with a vehicle dynamics model and a preset trajectory tracker, to obtain preset risk assessment data. Specifically, the preset driving data in this embodiment includes vehicle state information and environmental obstacle information. For example, for typical lane-changing and following scenarios, the present invention acquires the state vectors of the vehicle and surrounding obstacle vehicles in real time, including position, speed, and heading angle deviation. Correspondingly, the preset risk assessment data in this embodiment includes longitudinal collision risk data, lateral collision time data, and forward time data. The present invention assesses the longitudinal collision risk (TTC), lateral collision time (TTA), and forward time (TTF) based on the collected state vectors of the vehicle and surrounding obstacle vehicles. Accordingly, the longitudinal collision risk quantification in this embodiment is based on the real-time distance between the two vehicles. With speed difference The predicted rear-end collision time is shown in the following formula:
[0054] Where, in the formula yes The position coordinates of the vehicle ahead at any given time; yes The vehicle's position coordinates at all times; yes The longitudinal speed of the vehicle at all times; yes The longitudinal speed of the vehicle in front at any given moment.
[0055] Furthermore, in this embodiment of the invention, the lateral collision avoidance time quantification is based on the time required for the vehicle to move laterally to the minimum safe position to assess the risk of a lateral collision. Correspondingly, the forward time quantification is based on the remaining time threshold for the primary vehicle and the adjacent vehicle to pose a substantial collision threat in the longitudinal dimension.
[0056] Furthermore, this embodiment of the invention uses a nonlinear correction function to transform preset risk assessment data into comprehensive risk assessment indicators, and then determines safety benefit data based on these comprehensive risk assessment indicators. Specifically, this embodiment of the invention introduces a nonlinear correction function to transform the above indicators into comprehensive risk indicators. ,in Vertical risk coefficient: comprehensive consideration and The ratio relationship is used to enhance risk sensitivity through an exponential function. Among them, It is the lateral risk coefficient: based on the speed-related critical safety distance. actual distance The ratio is determined. Accordingly, embodiments of the present invention construct a total risk index. And derive the security sub-benefits. As shown in the following formula:
[0057] This return value reflects the current level of environmental risk in real time: the higher the risk, the lower the risk. The closer it is to 0; conversely, when the environment is absolutely safe, Approaching 1.
[0058] In some embodiments of the present invention, driving authority allocation analysis is performed through a shared control unit based on conflict index data and safety benefit data to obtain desired control authority allocation parameters, including but not limited to the following steps: Construct a pre-defined multi-dimensional sub-reward function. This pre-defined multi-dimensional sub-reward function includes a conflict sub-reward function, a safety sub-reward function, a participation sub-reward function, and a comfort sub-reward function.
[0059] Construct the total payoff function based on the conflict sub-payoff function, the safety sub-payoff function, the participation sub-payoff function, and the comfort sub-payoff function.
[0060] Based on conflict index data and security benefit data, the total benefit function is solved using Nash equilibrium to determine the expected control authority allocation parameters.
[0061] In this specific embodiment, the present invention first constructs a preset multidimensional sub-payoff function, and then constructs a total payoff function based on each sub-payoff function in the multidimensional sub-payoff function. Specifically, the present invention transforms the allocation of human-machine co-driving permissions into a game problem of dynamic interaction between two intelligent agents, and by introducing a control system as a "neutral arbitrator," solves the Nash equilibrium solution at each decision moment, thereby determining the desired control permission allocation parameters, such as... Figure 9 In this embodiment of the invention, the permission allocation model architecture quantifies the total control permission to 1 and defines the final front wheel steering angle output. Enter for driver Input to the autonomous driving system The weighted combination is shown in the following formula:
[0062] Where, in the formula This refers to the driver control authority coefficient that needs to be determined in real time in this embodiment of the invention.
[0063] Accordingly, such as Figure 10 As shown, in order to balance safety, intent compliance, and driving experience, this embodiment of the invention constructs a four-dimensional sub-benefit function: Conflict sub-benefits ( Based on the monitored ErrP conflict index If the driver's decision is deemed safe, then Encourage compliance with the driver's intentions; if dangerous, then... Forcefully reduce permissions.
[0064] Safety sub-benefits ( ): Output in real time through the collision risk model.
[0065] Participation in the benefits of the degree ( ): Defined by a nonlinear sine-arctangent combination function, it aims to avoid any single party from having complete control and to ensure that both humans and machines maintain a reasonable level of participation.
[0066] Comfort benefits ( ): By predicting lateral acceleration in the future time domain The value is then normalized to obtain the result; the closer the value is to 1, the smoother the trajectory.
[0067] Furthermore, in this embodiment of the invention, a weighted linear combination method is used to construct the total revenue function for both human and machine agents, as shown in the following equation:
[0068] Based on experimental calibration, the weights of each utility item were set as follows: safety weight. Conflict weights Participation weight Comfort weight This invention's embodiments maximize the Nash product. Solve for the optimal permission allocation coefficient at the current moment. This refers to the parameters for assigning control permissions.
[0069] It should be noted that, to ensure the implementation of the aforementioned monitoring and allocation algorithms, this embodiment of the invention constructs a complete "perception-decision-execution" closed-loop architecture. This architecture is implemented through a multi-software co-simulation platform and specifically includes: Perception layer: Utilizes Ideal Radar (AIR) and Imaging Sensor (TIS) to acquire real-time information on the road centerline, vehicle status, and surrounding obstacles.
[0070] Decision planning layer: Integrates conflict monitoring and game theory decision-making models. When the vehicle approaches a decision point, the system sends a visual warning signal to the driver via the Psychtoolbox module and simultaneously initiates the ErrP monitoring program.
[0071] Execution control layer: Lateral control is achieved by the LQR controller, which outputs the final front wheel steering angle based on the weights assigned in the game; longitudinal control uses a PID algorithm to adjust the throttle opening and braking pressure to ensure that the vehicle speed remains stable at the set target (e.g., 54 km / h).
[0072] Meanwhile, in response to different types of conflicts that may occur in actual driving, this invention exhibits different adaptive logic in the following three typical implementation scenarios: Safety conflict scenario: When the driver and the machine have different lane-changing intentions (such as overtaking on the left versus overtaking on the right), but both decisions are safe, the system identifies the higher participation benefit through game theory and allocates more permissions to the driver. (Higher), causing the trajectory to deviate from the driver's reference path, ensuring the driver's subjective sense of control.
[0073] Driver error scenario: When a driver intends to change lanes dangerously (e.g., into a lane with an obstacle), and the machine makes a safe decision, the physiological conflict detected by ErrP is combined with the safety sub-benefit. Due to the safety benefits... The threshold is extremely low; the game model will automatically reduce the driver's authority, allowing the machine to take the lead in obstacle avoidance and ensuring driving safety.
[0074] Machine error scenario: When the machine makes a dangerous decision due to perception limitations, while the driver's intention is correct, the low safety of the machine's decision will trigger the defense mechanism of the game model. The system will quickly transfer control to the driver and generate a lane change trajectory earlier and more smoothly than a single autonomous driving system through the anthropomorphic path replanning function.
[0075] Experiments show that, compared to a single manual driving mode, the positional error in trajectory tracking is significantly reduced in this embodiment of the invention, and the path tracking accuracy is improved by approximately 62.47%, effectively alleviating the path deviation problem caused by frequent driver fine-tuning. Simultaneously, by constraining lateral acceleration and its rate of change through a game theory model, the vehicle's lateral movement is smoother, and the measured extreme value of lateral acceleration is significantly lower than in conventional autonomous driving modes, improving ride comfort by approximately 49.87%. Correspondingly, this embodiment utilizes a brain-computer interface to monitor ErrP signals, achieving millisecond-level physiological conflict recognition. Compared to indirect inference methods relying on physical characteristics such as torque, this embodiment can more directly and accurately capture the driver's genuine sense of unease, providing a scientific feedback mechanism for permission allocation. Furthermore, the allocation scheme generated based on game theory allows for earlier and smoother lane changes compared to single autonomous driving, and the generated trajectory not only meets safety constraints but also better aligns with the driver's subjective operational preferences, contributing to enhanced human-machine trust. In addition, the construction of the multi-criteria benefit function in the embodiments of the present invention ensures that the system can automatically adjust permissions under extreme conflict conditions (such as human / machine error), which can respect human participation intentions when it is safe and achieve autonomous takeover when it is dangerous, thus having strong engineering application value.
[0076] It is readily understood that the embodiments of the present invention can effectively alleviate the problem of subjective and objective fusion in human-machine conflict recognition. By introducing brain-computer interface technology to monitor error-related potentials (ErrP), it alleviates the perceptual lag caused by relying solely on physical features to infer intent, and achieves real-time quantitative feedback in the physiological dimension. At the same time, it alleviates the imbalance problem of dynamic allocation of permissions under multiple constraints. It uses game theory to construct an allocation model and solves it through Nash equilibrium to balance security, intent compliance, and comfort, and avoids trajectory oscillation and trust reduction caused by permission switching. Furthermore, it solves the problem of the robustness of control strategies to complex conflict scenarios. By integrating the benefits of multiple criteria to design a shared control scheme with a physiological feedback closed loop, it enables vehicles to achieve high-precision trajectory tracking and smooth decision-making under various human-machine deviation conditions.
[0077] Please see Figure 11 This application also provides a human-machine co-driving and shared control device that can implement the above-mentioned method. The device includes: The first module 210 is used to acquire the driver's brainwave signal data through a preset brain-computer interface, and then analyze the brainwave signal data through a human-machine conflict monitoring unit to determine the conflict index data.
[0078] The second module 220 is used to acquire preset driving data, and then use the vehicle safety status assessment unit to assess the vehicle collision risk of the preset driving data to obtain safety benefit data.
[0079] The third module, 230, is used to analyze driving authority allocation through the shared control unit based on conflict index data and safety benefit data, to obtain the desired control authority allocation parameters. The shared control unit is constructed based on bargaining game theory.
[0080] The fourth module 240 is used to plan a path and generate a target driving path based on the desired control authority allocation parameters and preset reference paths. The preset reference paths include a driver reference path and an autonomous driving reference path.
[0081] It is understood that the content of the above method embodiments is applicable to the present device embodiments. The specific functions implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0082] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.
[0083] It is understood that the content of the above method embodiments is applicable to this device embodiment. The specific functions implemented by this device embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0084] Please see Figure 12 , Figure 12 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 310 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 320 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 320 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 320 and is called and executed by the processor 310 using the methods described in the embodiments of this application. Input / output interface 330 is used to realize information input and output; The communication interface 340 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 350 transmits information between various components of the device (e.g., processor 310, memory 320, input / output interface 330, and communication interface 340); The processor 310, memory 320, input / output interface 330 and communication interface 340 are connected to each other within the device via bus 350.
[0085] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.
[0086] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.
[0087] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.
[0088] It is understood that the content of the above method embodiments is applicable to the embodiments of this program product. The specific functions implemented by the embodiments of this program product are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0089] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
[0090] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.
[0091] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.
[0092] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.
[0093] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.
[0094] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0095] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.
[0096] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0097] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0098] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0099] If the integrated unit is implemented as 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 this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0100] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.
Claims
1. A human-machine co-driving and shared control method, characterized in that, The method includes the following steps: The driver's brainwave signal data is acquired through a preset brain-computer interface, and then the brainwave signal data is analyzed by a human-machine conflict monitoring unit to determine the conflict index data. Preset driving data is acquired, and then the vehicle safety status assessment unit performs a vehicle collision risk assessment on the preset driving data to obtain safety benefit data. Based on the conflict index data and the safety benefit data, driving authority allocation analysis is performed through the shared control unit to obtain the desired control authority allocation parameters; wherein, the shared control unit is constructed based on bargaining game theory; Based on the desired control permission allocation parameters and the preset reference path, path planning is performed to generate a target driving path; wherein, the preset reference path includes a driver reference path and an autonomous driving reference path.
2. The method according to claim 1, characterized in that, The process of acquiring the driver's electroencephalogram (EEG) signal data through a preset brain-computer interface, and then analyzing the EEG signal data through a human-machine conflict monitoring unit to determine conflict index data, includes: Error-related potential data is acquired through the preset brain-computer interface, and conflict feature data is then extracted from the error-related potential data. The conflict feature data is identified by a preset machine learning model, and then the conflict quantization calculation is performed based on the identified conflict frequency band signals to obtain the conflict index data.
3. The method according to claim 2, characterized in that, The step of extracting conflict feature data from the error-related potential data includes: The error-related potential data is filtered by a preset bandpass filter to obtain the first EEG signal; The first EEG signal was processed by an independent component analysis algorithm to remove artifacts, resulting in the second EEG signal. A spatial filter is constructed using a common spatial pattern algorithm, and then the spatial filter is used to extract spatial features from the second EEG signal to obtain the conflict feature data.
4. The method according to claim 1, characterized in that, Before performing the steps of acquiring preset driving data and then using a vehicle safety status assessment unit to assess the vehicle collision risk of the preset driving data to obtain safety benefit data, the method further includes: Construct a vehicle dynamics model; wherein the vehicle dynamics model includes air resistance and environmental disturbance terms; A preset trajectory tracker is constructed based on the linear quadratic regulator algorithm.
5. The method according to claim 4, characterized in that, The process of acquiring preset driving data and then using a vehicle safety status assessment unit to assess the vehicle collision risk based on the preset driving data to obtain safety benefit data includes: Acquire the preset driving data; wherein the preset driving data includes vehicle status information and environmental obstacle information; Based on the vehicle status information and the environmental obstacle information, a collision risk analysis is performed using the vehicle dynamics model and the preset trajectory tracker to obtain preset risk assessment data; wherein, the preset risk assessment data includes longitudinal collision risk data, lateral collision time data, and forward time data; The preset risk assessment data is transformed into a comprehensive risk assessment index through a nonlinear correction function, and then the safety benefit data is determined based on the comprehensive risk assessment index.
6. The method according to claim 1, characterized in that, The step of analyzing driving authority allocation through a shared control unit based on the conflict index data and the safety benefit data to obtain desired control authority allocation parameters includes: Construct a preset multidimensional sub-benefit function; wherein, the preset multidimensional sub-benefit function includes a conflict sub-benefit function, a safety sub-benefit function, a participation sub-benefit function, and a comfort sub-benefit function; Construct a total benefit function based on the conflict sub-benefit function, the safety sub-benefit function, the participation sub-benefit function, and the comfort sub-benefit function; Based on the conflict index data and the security benefit data, the total benefit function is solved using Nash equilibrium to determine the expected control authority allocation parameters.
7. A human-machine co-driving and shared control device, characterized in that, The device includes: The first module is used to acquire the driver's brainwave signal data through a preset brain-computer interface, and then analyze the brainwave signal data through a human-machine conflict monitoring unit to determine the conflict index data. The second module is used to acquire preset driving data, and then use the vehicle safety status assessment unit to assess the vehicle collision risk of the preset driving data to obtain safety benefit data. The third module is used to perform driving permission allocation analysis through the shared control unit based on the conflict index data and the safety benefit data to obtain the desired control permission allocation parameters; wherein, the shared control unit is constructed based on bargaining game theory; The fourth module is used to perform path planning based on the desired control permission allocation parameters and preset reference paths to generate a target driving path; wherein the preset reference paths include a driver reference path and an autonomous driving reference path.
8. An electronic device, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method as described in any one of claims 1 to 6.
9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 6.
10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 6.