3202 results about "Cockpit" patented technology

An attitude display system for use in an aircraft has at least one display unit visible by a pilot within a cockpit. A symbology generator is configured for receiving input from at least one attitude sensor and generating symbology image data representative of a pear-shaped attitude indicator, distorted vertically to represent a pitch of the aircraft and rotated to represent a roll angle of the aircraft. The symbology generator may further be configured for adding a horizontal line that represents a horizon, a regular trapezoid representing a downward gravity vector at a base of the trapezoid, a pitch angle of the aircraft, split wings representing an angle of attack of the aircraft, and a stall warning in a sawtooth crack shape. A tail portion of a T-shape with the curved top edge is itself curved in a direction opposite that of an aircraft spin direction.

The invention discloses a cockpit active recommendation system and method based on a knowledge graph and semantic reasoning, and relates to the technical field of intelligent cockpits. The system receives natural language voice input of a user, executes voice recognition and semantic analysis, extracts user intention, keywords and slot entities, generates structured semantic information, constructs or calls a knowledge graph structure with semantic relation edges in combination with environment context information, and obtains the knowledge graph structure with the semantic relation edges. Semantic path reasoning is carried out based on the path dependence weight and the semantic similarity, a semantic edge label guided graph attention mechanism is introduced to calculate a path consistency score, a candidate recommendation set is generated, the semantic fitting degree and the path score are fused to sort and output recommendation content, and the graph edge weight and the user portrait are updated based on user feedback. According to the method, semantic understanding precision, recommendation path interpretability and system adaptive capacity are improved, and the method is suitable for personalized voice recommendation, man-machine interaction and scene linkage control tasks in an intelligent cockpit.

The invention discloses a vehicle seat self-adaptive adjusting method and system based on environment perception and a medium. The method comprises the steps that physiological state data and human body image data of a target passenger and cabin environment data of a target cabin are obtained; determining body surface temperature and humidity, heart rate parameters and human body metabolic rate of the target passenger according to the physiological state data, and identifying clothing thermal resistance parameters of the target passenger according to the human body image data; determining a human body thermal comfort index of the target cabin according to the human body metabolic rate, the clothing thermal resistance parameter and the cabin environment data; inputting the body surface temperature and humidity, the heart rate parameter and the human body thermal comfort index into a pre-trained thermal comfort demand prediction model to obtain a target thermal comfort demand of the target passenger; and performing heating adjustment and / or ventilation adjustment on the seat of the target cabin according to the target thermal comfort demand. The riding comfort and riding experience of the user are improved, and the method can be widely applied to the technical field of vehicle control.

The invention discloses an intelligent cabin interaction method and system combined with digital twinning, and relates to the technical field of man-machine interaction, and the method comprises the steps: collecting the multi-dimensional perception data of a cabin, and constructing a layered cabin digital twinning model; performing scene analysis on the cockpit digital twin model, extracting driver state features, predicting driver potential interaction demands, generating a multi-modal interaction strategy set in combination with the driver potential interaction demands and a current driving scene, and performing simulation evaluation on the multi-modal interaction strategy set to obtain an interaction evaluation index set; performing multi-objective decision optimization under the driving safety constraint based on the interaction evaluation index set, and determining and adaptively executing an optimal interaction strategy; and acquiring actual execution data and driver feedback data of the optimal interaction strategy, and incrementally updating the hierarchical cabin digital twin model. According to the invention, the interaction adaptability and the safety reliability of the intelligent cabin system in a complex driving scene are obviously improved, and an effective technical support is provided for realizing intelligent man-machine cooperative driving.

The invention relates to the technical field of fire fighting systems, in particular to an intelligent cockpit system for fire fighting, which comprises a perception analysis layer, a decision processing layer, a command execution layer and a reinforcement learning closed-loop architecture for mixed reward shaping, and integrates video streams, audio communication and firefighter physiological data containing heart rate variability through a deep multi-modal fusion module. Generating a global fire scene situation of physical constraint verification; through a risk sensitive type three-dimensional fire scene deduction module, a fire extinguishing strategy considering efficiency and safety is generated; through an immersive augmented reality visual interface, in combination with a synchronous positioning and mapping technology, precise navigation in a complex environment and superposed display of situation information containing risk levels are realized; a Q learning algorithm continuous optimization strategy including domain knowledge intermediate process rewards is adopted, closed-loop optimization of perception-analysis-decision-execution-feedback is achieved, and therefore the overall efficiency of fire rescue operation in modern complex disaster scenes is remarkably improved.

The invention discloses an aircraft cabin environment personalized adjustment method based on sentiment analysis, and the method comprises the steps: collecting the facial expression, voice waveform and environment parameters of a passenger in real time through a cabin multi-source sensor, and generating a standardized physiological signal matrix and anonymized voice features through noise reduction and feature extraction; inputting the physiological signal matrix and the voice features into a pre-trained deep learning model, outputting an emotion index and a classification label, and updating model parameters through a federal learning framework; dynamically generating temperature, humidity and oxygen concentration adjusting instructions and dynamic weights by adopting a fuzzy reasoning system in combination with the emotion indexes and passenger preset preferences; environment adjustment is executed through a closed-loop control system, and environment parameter errors are fed back; synchronously updating a fuzzy inference system rule base and deep learning model parameters by utilizing reinforcement learning in combination with environmental parameter errors and emotion index changes, and completing optimization of a closed loop; according to the invention, real-time dynamic regulation and control and continuous adaptation optimization of the personalized cabin environment can be realized.

The invention provides a multi-screen cabin entertainment system, and relates to the technical field of embedded systems and intelligent cabins, and the multi-screen cabin entertainment system is characterized in that an application layer is deployed with a plurality of cabin entertainment applications, and when the application layer is started, content rendering requests are sent out respectively; the hardware layer is provided with a development board, is provided with a CPU core cluster and a coprocessor, and is connected with different cabin display screens through different video output interfaces. The system layer respectively distributes an independent thread and an independent frame buffer area for each content rendering request on the CPU core cluster, and establishes a mapping relation between each frame buffer area and each video output interface; and sending operation result data of each independent thread to a coprocessor for rendering, correspondingly storing a rendering result in a frame buffer area, and then sending the rendering result to a cabin display screen for display through a video output interface. The method has the advantages that on the premise that the number of hardware coprocessors is not increased, complete separation of multi-screen rendering threads and frame buffer is achieved, and multiple cockpit entertainment applications can run at the same time and do not interfere with one another.

The invention discloses a cockpit scene recommendation method and device, a storage medium, a program product and a vehicle, and relates to the technical field of artificial intelligence, and the method comprises the steps: carrying out the preprocessing of man-machine conversation text data of a voice burying point log of the vehicle, obtaining the natural language description of a user, carrying out the similarity screening in a preset scene knowledge base, and obtaining a voice burying point log of the vehicle; obtaining corresponding scene knowledge; splicing the user natural language description with the scene knowledge to obtain a scene recommendation prompt word; and performing scene action reasoning on the scene recommendation prompt word through the large model agent, and outputting a scene action execution list corresponding to the scene knowledge. According to the method, man-machine conversation text data is converted into user natural language description, corresponding scene knowledge is screened, scene action reasoning is carried out based on the powerful text understanding and generating capacity of a large model agent, and the intelligent scene recommendation service of the cockpit can be automatically realized according to the voice description of a vehicle owner; and the accuracy of cabin scene recommendation is improved.

The invention relates to the technical field of cabin control, and discloses an audio parameter optimization method and device for a vehicle cabin, and a vehicle terminal. According to the application, modeling is carried out on a vehicle cabin to obtain a cabin acoustic model, and the cabin acoustic model is driven to carry out acoustic simulation on the vehicle cabin through cabin audio parameters corresponding to a playing source area, so that target acoustic parameters corresponding to a target area are determined according to an acoustic simulation result; according to the method, the audio parameters of the cockpit are optimized according to the target acoustic parameters, the audio parameters are set by adopting unified parameters compared with different vehicle cockpits, an acoustic model is obtained through modeling so as to simulate different vehicle cockpits, and then the target acoustic parameters obtained through acoustic simulation guide the audio parameters of the cockpit to be optimized. Therefore, the flexibility of audio parameter optimization is improved, the optimization difficulty of the cabin audio parameters is reduced, and the user experience is improved.

The invention belongs to the field of eye movement data tracking, particularly relates to a pilot eye movement tracking method, system and equipment based on multi-camera fusion and a storage medium, and aims to solve the problems that in the prior art, the visual angle of a traditional eye tracker is limited, and sight estimation errors are caused by head movement. The method comprises the following steps of: constructing a multi-view visual perception network by deploying multiple cameras and eye trackers in a cockpit; multi-source data space-time alignment is achieved through space coordinate system mapping and time synchronization, and head motion errors are compensated by combining dynamic cutting of eye interested areas and three-dimensional head posture estimation. The geometric path generates a physical driving vector based on iris fitting and an eyeball model, and the data path regresses the vector through a deep learning model; and converting to a world coordinate system through a head posture after fusion. High-precision data of the eye tracker is preferentially adopted in the central area, and field-of-view data of the system is expanded in the edge area. According to the invention, the tracking precision under field angle coverage and motion interference is obviously improved.

The invention discloses a pilot manipulation ability evaluation method and system based on multi-modal data and online learning driving. The method comprises the steps of determining an input parameter set, establishing a flight vision time sequence fusion network, and extracting each feature of the parameter set. And completing feature fusion of the multi-modal data through an attention mechanism and global aggregation. And constructing a pilot maneuvering ability evaluation architecture, and generating an offline pilot maneuvering ability evaluation model by designing a loss function based on a comprehensive scoring error and sub-scoring consistency constraint. Flight data and eye movement data streams generated in real time are fragmented and preprocessed by using a sliding window mechanism, a lightweight model based on pruning and parameter quantification technologies is designed for real-time scoring requirements, and feature expression ability in an offline pilot manipulation ability evaluation model is migrated to a lightweight online evaluation model. The method can be used for improving a pilot training scheme, optimizing the man-machine interface design of a cockpit, and improving the efficiency and safety of flight operation.

The invention relates to the technical field of TEM models, provides a pilot dynamic evaluation method, system and device based on a TEM model and a storage medium, and solves the problems of low accuracy of flight training evaluation and poor adaptability of a training scheme. The method comprises the steps that flight control, physiological monitoring and cockpit voice data are acquired, and a multi-modal data stream is formed through time synchronization; performing threat perception, error management and non-technical skill three-level evaluation on the feature vector by using a TEM model, generating corresponding indexes, and fusing the indexes into a comprehensive feature vector; respectively processing the time sequence and cognitive features by means of a dual-channel long-short-term memory network, and extracting deep features; performing decision tracing by adopting an SHAP value so as to locate a capability defect node; finally, defect types are matched through a personalized improvement scheme recommendation mechanism, and self-adaptive training content is generated. According to the invention, the accuracy of flight training evaluation and the adaptability of the training scheme are improved.

The invention provides a vehicle-mounted oxygen production control method and device, a vehicle-mounted oxygen production system and electronic equipment, and relates to the technical field of vehicle control. The method comprises the steps that the oxygen partial pressure in a cockpit, the altitude of the environment where the cockpit is located and the oxyhemoglobin saturation of passengers in the cockpit are obtained; determining an oxygen supply demand value based on the oxygen partial pressure, the altitude and the oxyhemoglobin saturation; and based on the oxygen supply demand value, an oxygen generation module is controlled to start oxygen generation or stop oxygen generation. The starting and stopping of the oxygen generation module are controlled based on the oxygen supply demand value, waste caused by starting of the oxygen generation module when oxygen supply is not needed can be avoided, and therefore the energy consumption of vehicle-mounted oxygen generation can be greatly reduced.

An aircraft includes a flight deck including a flight deck door configured to be moved between an open position and a closed position, a first monument, second monument, and a secondary barrier door moveably secured between the first monument and the second monument. The secondary barrier door is configured to fold into a stowed position in which a path to the flight deck is open, and unfold into a deployed position in which the path to the flight deck is closed.

The invention discloses a cockpit visual work efficiency evaluation method. The method comprises the following steps: establishing a cockpit geometric model according to a real cockpit; establishing a cockpit light environment model according to the cockpit light environment parameters; establishing a cockpit optical material model according to the optical material parameters of the inner and outer surfaces of the cockpit; fusing and establishing a cockpit optical model; a simulation detector is arranged on the cockpit optical model; according to the cockpit optical model and the analog detector, human eye vision optical simulation and dazzle source space tracking and positioning are carried out; according to a glare source space tracking and positioning result, a cockpit structure is optimized; and performing cockpit visual work efficiency evaluation according to a cockpit structure optimization result. According to the cockpit visual work efficiency evaluation method, reasonable evaluation of cockpit visual work efficiency readable visibility in various environments is achieved, simulation is conducted on the cockpit light environment through computer software, and adjustment can be conducted according to the working habit, the vision condition, the age and other factors of each pilot.

The invention provides a low-altitude typical scene-oriented aircraft cockpit simulation method, system and device and a storage medium, and the method comprises the steps: constructing a dynamic extensible low-altitude scene library through fusing mixed reality (MR) and a high-precision digital twin model; in combination with a man-machine in-the-loop multimode interaction technology (eye movement-gesture-voice cooperation), the operation naturalness is improved; and an intelligent optimization algorithm driven by reinforcement learning is introduced, adaptive adjustment of flight control parameters is realized, and finally a'perception-decision-evaluation 'closed-loop training system is formed. According to the invention, digital twinning, mixed reality (MR), multi-modal sensing and adaptive control technologies are fused, pilot low-altitude flight skill training, man-machine work efficiency evaluation and intelligent aircraft control system verification are carried out, and the method can be expanded to UAM operator training and unmanned aircraft simulation testing. The technical barrier of low-altitude flight skill training at present is solved, and the technical blank of low-altitude dynamic scene simulation is filled up.

The invention provides a cockpit multi-tone area management method, a related device and a communication system. The cabin system may angry the plurality of screens into a plurality of ride zones according to a profile. Each riding area can be associated with a sound area. The cabin system can dynamically create and delete the sound area according to the access and removal of the sounding device. The cabin system can perform unified management on audio focus, volume, audio routing and media session in a sound area. The application corresponding to one driving area can apply for the audio focus in the sound area associated with the driving area, and the sound production equipment in the sound area is used for playing the audio. Audio playing in different sound areas can be independent and does not affect each other. Therefore, different users in the vehicle can use different sound areas to listen to the audio conveniently, and the audio listening experience of the users in the vehicle is improved.

The invention discloses an intelligent cockpit driver intention prediction method and system fusing scene semantics, and the method comprises the steps: collecting cockpit interaction data and driving dynamic data, extracting a data sequence of a specified time length through a sliding window method, and constructing a driver interaction intention data set; constructing an instruction fine tuning data set of scene semantic understanding based on logic rules and manual inspection, and converting the driving dynamic data into a scene semantic text through a fine tuning large language model; training a deep learning model by using the constructed driver intention data set, and outputting the most probable driver interaction intention; according to the method, the limitation of independent modeling of a driver action sequence and spatio-temporal information in a traditional method is effectively solved, collaborative modeling of driver interaction actions and scene spatio-temporal characteristics is formed by fusing context information of a scene spatio-temporal causal relationship, and accurate prediction of interaction intentions of various services such as air conditioners, music and seats in an intelligent cabin is achieved.

The invention provides a method and a system for generating deviation guide information in an approaching landing section of an aircraft and a readable storage medium, and the method comprises the steps: S1, real image collection and processing: capturing an outboard environment image through an airborne visual sensor, and extracting key feature points of a target landing airport runway; s2, virtual image generation: based on the airport three-dimensional high-precision map, generating an approach glide image under a visual angle of a virtual camera; s3, image feature registration: carrying out spatial alignment on the key feature points of the runway in the real image and the key feature points of the runway in the virtual image; the pixel deviation of the runway center point in the real image and the virtual image is calculated, glide guidance is provided for a pilot or is directly input into an automatic flight system.The method can be used for the aircraft in the approaching landing section, the enhanced outboard visual display is provided for the pilot in the cockpit based on the equivalent ILS deviation guidance method, and the visual guidance effect is improved. And providing approaching glide deviation guide information.

The invention relates to the technical field of vehicle testing, in particular to a simulation testing method and device for an intelligent cockpit, and the method comprises the steps: carrying out the simulation environment configuration of a cockpit of a to-be-tested vehicle, and obtaining a training set according with the cockpit; performing data acquisition on the internal scene of the cockpit by adopting a camera to obtain an internal scene image of the cockpit; adding the training set to the internal scene image, inputting the test model, and identifying the training set to obtain an identification result; and performing response based on the identification result, controlling the internal scene of the to-be-tested vehicle, and performing verification optimization on the to-be-tested vehicle. The simulation platform can simulate various complex and extreme driving scenes and illumination environments, provide comprehensive test coverage, generate a training set in a virtual environment, test the performance of an internal system of the intelligent cockpit under the invasion, continuously optimize and improve the system, and improve the resistance to the invasion and the overall safety of the system.

The invention discloses a function recommendation method and system based on an intelligent cabin, a vehicle and equipment. The function recommendation method based on the intelligent cockpit comprises the following steps: acquiring a user identity; when the user identity is determined as a target user, obtaining a driving scene of a vehicle; the driving scene is input into a pre-trained recommendation model, function state selection matched with the driving scene is obtained, and the recommendation model is pre-trained according to collected operation data of a target user in a non-driving scene; and recommending the function state selection to the target user. By the adoption of the automobile seat cushion, the degree of adhesion with a driver is higher, the automobile seat cushion is closer to an ideal seat cabin, meanwhile, the driver can drive more securely, the maximum relaxation feeling is met, the purpose of reducing driving fatigue is achieved, customers can be better served, and the individualized requirements of more scenarios can be met.

The invention provides a split type hovercar which comprises an aircraft module, a cabin module and a chassis module, and the split type hovercar has a ground mode, a flight mode, a take-off mode and a landing mode. The aircraft module is connected with the top of the cabin module, and the chassis module is detachably connected with the bottom of the cabin module; wherein after the split type hovercar is switched from a take-off mode to a flight mode, the chassis module is separated from the cabin module; after the split type hovercar is switched from the flight mode to the landing mode, the chassis module is in butt joint with the cabin module. Through the arrangement, after the split type hovercar is switched from a take-off mode to a flight mode, the chassis module can be separated from the cabin module, so that the flight resistance is reduced, and the flight energy consumption is reduced; after the split type hovercar is switched from the flight mode to the landing mode, the chassis module can be in butt joint with the cabin module, so that the split type hovercar slides and lands stably, and the landing efficiency is improved.

The invention discloses an intelligent cabin personalized recommendation method combining environmental perception and passenger preference, and relates to the technical field of intelligent automobiles. Comprising the following steps: data acquisition: based on environment feature big data, vehicle intelligent cabin passenger preference data and historical recommendation schemes, collecting environment data, passenger preference data and historical recommendation schemes to obtain environment data items, passenger preference data items and historical recommendation scheme data items; and data analysis: based on the environment data item, the passenger preference data item and the historical recommendation scheme data item, analyzing the environment data item, the passenger preference data item and the historical recommendation scheme data item to obtain an analysis result. Through environment feature recognition and emotion preference recognition, the current driving scene of the passenger can be accurately recognized, various factors such as the vehicle speed, the road condition information, the weather condition and the geographic position are considered, and driving scene recognition is more accurate and comprehensive.

The invention relates to an intelligent cabin man-machine interaction evaluation method, device and equipment and a medium, and the method comprises the steps: judging whether there is an evaluation demand, and the evaluation demand comprises to-be-tested environment building data and a user interaction requirement; if the evaluation requirement exists, building a virtual scene according to the to-be-tested environment building data, and collecting simulator state data, virtual environment data and multi-modal data of a user when the user performs man-machine interaction in the virtual scene based on the user interaction requirement; and obtaining a subjective evaluation result of the user, obtaining an objective evaluation result according to the simulator state data, the virtual environment data and the multi-modal data, and obtaining a man-machine interaction evaluation result according to the subjective evaluation result and the objective evaluation result. Therefore, the problems of poor flexibility, evaluation lagging, high iteration cost, high evaluation subjectivity and the like of an evaluation means based on a physical model machine are solved, the flexibility, the real-time performance and the objectivity of evaluation are improved, and the evaluation efficiency is improved.

The invention discloses a domain control optimization method and system based on cockpit internal environment recognition, and relates to the field of domain control optimization, and the method comprises the steps: building a dynamic correlation model of the change of real-time environment parameters outside a cockpit and the expected change of environment parameters in the cockpit; constructing a pre-adjustment optimization feedback loop, and carrying out feedback closed-loop primary adjustment on the environmental parameters in the cabin; performing unified modeling on the historical multi-source environment parameters, and establishing an environment situation recognition model in the cabin; based on the identified environment situation type and the corresponding historical data of the human physiological state, establishing an association relationship between the situation type and the human physiological state; the method comprises the following steps: predicting a human physiological state based on expected environmental parameter prediction in a cabin, environmental situation recognition and a correlation between a situation type and the human physiological state, adaptively generating a domain control strategy by screening historical domain control strategy records, and carrying out secondary adjustment on environmental parameters in the cabin.

The invention belongs to the technical field of voice processing, and discloses an intelligent cockpit multi-mode voice interaction system and method, and the system comprises a voice triggering unit which collects the environment audio and video information in a cockpit, judges whether to enter a voice interaction mode or not through combining with the environment perception parameters in a vehicle, and sends the voice interaction mode to the vehicle; when the voice interaction triggering condition is satisfied, generating a voice interaction input signal matched with the current environment; the mouth shape analysis unit is used for carrying out acoustic feature extraction on the voice interaction input signal, synchronously analyzing the lip motion trail of the driver in the video information, establishing a corresponding relation between voice phonemes and mouth shape motion, and forming a joint analysis feature; the candidate generation unit is used for carrying out segmented alignment on the joint analysis features and constructing a continuous multi-modal fragment sequence; performing time synchronization on the multi-modal fragment sequence, and projecting the multi-modal fragment sequence to a predefined intention space to obtain a candidate intention set containing different candidate intentions; and the man-machine interaction experience of the intelligent cabin is improved.

The invention relates to the technical field of power system management, and discloses a cockpit multi-scene switching output control method based on artificial intelligence, which comprises the following steps: S1, real-time data acquisition and fusion: acquiring power load, equipment state and environment data through a power load sensor, an equipment state sensor and an environment data sensor, fusing the collected data and constructing a multi-dimensional state space to represent a current power system state; and S2, deep reinforcement learning and optimal decision path calculation: constructing a state space by using the collected real-time data, and selecting a most suitable display mode through a reinforcement learning training model to ensure that the system can make an optimal decision in different scenes. According to the multi-scene switching output control method driven by artificial intelligence, intelligent and dynamic adjustment of the display content is realized, the system response speed, the decision efficiency and the stability are improved, and the information display precision and the operator decision support capability in the power dispatching process are improved.

The embodiment of the invention discloses a cabin control method and device, vehicle-mounted equipment and a computer program product, and the method comprises the steps: obtaining multi-modal data of a user in a cabin of a vehicle, inputting the multi-modal data of the user into a current neural network model, generating a cabin service, and executing a cabin function corresponding to the cabin service, the current neural network model is obtained by performing local training on a preset neural network model by using historical data of the user in the cabin, and the historical data of the user is locally generated multi-modal data of the user and cabin service corresponding to the locally generated multi-modal data of the user. Therefore, each data in the multi-modal data is considered in the obtained model, so that more intelligent cockpit service is provided for users.

The invention discloses a duplex-screen intelligent cabin interaction method and system, and relates to the technical field of intelligent cabin interaction, and the method comprises the steps: collecting user operation behavior data, vehicle state data and environment perception data in real time through a vehicle-mounted sensor group, a man-machine interaction interface and a vehicle bus; establishing a multi-dimensional user data model containing an interaction feature vector, a driving mode matrix and a preference configuration tree; the method comprises the following steps: analyzing user historical interaction data based on an adaptive weighting algorithm, and calculating use frequency weight coefficients of different functional modules; according to the duplex-screen intelligent cockpit interaction method and system provided by the invention, the precision of a user portrait is remarkably improved by constructing a multi-dimensional user data model, personalized interaction experience is realized, and the interface layout and interaction logic can be dynamically adjusted according to a real-time driving scene by adopting a self-adaptive weighting algorithm and a random forest classifier, so that the user experience is improved. The real-time performance and the adaptability of the system are enhanced, and the interaction experience is optimized.

The invention discloses a real-time emotion perception and voice interaction system for an intelligent cabin. The system comprises a multi-modal data acquisition module used for synchronously acquiring a facial image, a voice signal and text input of a driver; the visual feature enhancement unit is used for carrying out restoration and emotion distribution extraction on the low-quality image; the audio noise reduction and feature extraction unit is used for extracting voice emotion features; the text emotion coding unit is used for fusing relative position coding and context semantic information; the cross-modal fusion module is used for outputting an emotion classification result by integrating visual, audio and text features; the personalized emotion database is used for storing historical emotion data of the user and performing emotion trend prediction and early warning judgment; the large language model feedback module is used for generating structured cue words according to the emotion recognition result and the driving situation and generating natural language feedback; the voice synthesis and output module is used for adjusting voice parameters and performing feedback output through a vehicle-mounted multi-channel; according to the invention, the emotion expression of human-vehicle interaction is enhanced.