CN-121982954-A - Pilot state monitoring method and related device based on unmanned aerial vehicle flight simulation

CN121982954ACN 121982954 ACN121982954 ACN 121982954ACN-121982954-A

Abstract

The invention discloses a pilot state monitoring method and a related device based on unmanned aerial vehicle flight simulation, and belongs to the technical field of aviation flight training; the method comprises the steps of configuring a simulation environment comprising a typical task scene and a preset fault scene of an unmanned aerial vehicle, synchronously collecting flight data, pilot operation data and pilot physiological data when a pilot trains in the simulation environment, triggering fault simulation in preset time, recording identification, diagnosis and treatment operation process data of the pilot on faults, aligning and fusing the collected data according to time axes, analyzing the operation performance, physiological state and psychological cognition of the pilot based on the fused data, and generating a comprehensive evaluation report. The method can realize comprehensive and quantitative evaluation of the workload, situational awareness and man-machine interaction efficiency of the unmanned aerial vehicle pilot. The method can be used for pilot selection training, human-computer interface evaluation and task optimization.

Inventors

WANG HAN
Liu sanli
WANG WENZHI
LIU FENGZHOU
LI LAIXI

Assignees

中国人民解放军空军军医大学

Dates

Publication Date: 20260505
Application Date: 20260202

Claims (10)

1. The pilot state monitoring method based on unmanned aerial vehicle flight simulation is characterized by comprising the following steps of: configuring a simulation environment comprising a typical task scene and a preset fault scene of the unmanned aerial vehicle; when a pilot trains in the simulation environment, synchronously acquiring flight data, pilot operation data and pilot physiological data; Triggering fault simulation under preset time, and recording the identification, diagnosis and treatment operation process data of the pilot on the fault; And aligning and fusing the collected flight data, pilot operation data, pilot physiological data and pilot fault identification, diagnosis and treatment operation process data according to time axes, and analyzing the pilot operation performance, physiological state and psychological cognition based on the fused data to generate a comprehensive evaluation report.
2. The pilot status monitoring method based on unmanned aerial vehicle flight simulation of claim 1, wherein the preset fault scenario comprises at least one of engine degradation, air parking, control surface blockage, sensor failure, or data link interruption.
3. The method of claim 1, wherein the pilot physiological data includes at least heart rate, heart beat interval, gaze point coordinates, pupil diameter, and blink frequency.
4. The pilot status monitoring method based on unmanned aerial vehicle flight simulation according to claim 1, wherein the operation performance analysis process specifically comprises: calculating fault identification time, wherein the fault identification time is from the moment of occurrence of the fault to the time of first operation of an alarm area or the time of mentioning the fault; Judging the correctness of the decision, namely judging whether a correct emergency program is started in a reasonable time; operational efficiency is calculated, expressed as the ratio of invalid operations during the treatment period.
5. The pilot state monitoring method based on unmanned aerial vehicle flight simulation according to claim 1, wherein the physiological state analysis process specifically comprises: Heart rate response calculating the rate of increase of the average heart rate over the baseline period during stress response; Visual attention distribution the percent residence time of the stress phase gaze point on each interface element was analyzed.
6. The pilot state monitoring method based on unmanned aerial vehicle flight simulation according to claim 1, wherein the psychological cognitive analysis process specifically comprises: Psychological burden calculating psychological burden index of decision treatment period according to pupil diameter and heart rate; And (3) a situational awareness score, namely after the fault is triggered, suspending the simulation and presenting a probe question to the pilot before the simulation is finished, and evaluating the situational awareness of the pilot according to the correctness and the response time of the probe question answer.
7. A system for implementing the pilot condition monitoring method based on unmanned aerial vehicle flight simulation of any one of claims 1 to 6, comprising: the unmanned aerial vehicle flight simulation module is used for simulating a flight dynamics model of the unmanned aerial vehicle and injecting system faults; the task planning and monitoring simulation module is used for providing a man-machine interaction interface for task planning and flight monitoring and recording an operation log; The flight visual simulation module is used for generating a visual synchronous with the state of the unmanned aerial vehicle and visualizing tasks and fault information; And the pilot physiological and psychological monitoring module is used for collecting physiological signals of the pilot and evaluating psychological load and situational awareness of the pilot.
8. The unmanned aerial vehicle flight simulation-based pilot condition monitoring system of claim 7, wherein the pilot physiological and psychological monitoring module comprises: the physiological signal acquisition unit is used for acquiring physiological data of pilots; And the psychological assessment unit is used for calculating psychological load index and assessing situational awareness through the probe problems.
9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the computer program, carries out the steps of a pilot condition monitoring method based on unmanned aerial vehicle flight simulation according to any one of claims 1-6.
10. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of a pilot condition monitoring method based on unmanned aerial vehicle flight simulation according to any one of claims 1-6.

Description

Pilot state monitoring method and related device based on unmanned aerial vehicle flight simulation Technical Field The invention belongs to the technical field of aviation flight training, and relates to a pilot state monitoring method and a related device based on unmanned aerial vehicle flight simulation. Background Unmanned aerial vehicle (Unmanned AERIAL VEHICLE, UAV) systems have been widely used in a number of fields such as military reconnaissance, civil mapping, logistics transportation, emergency rescue, and the like. With the continuous complicating of application scenes, higher requirements are put forward on the cognitive ability, emergency handling ability and man-machine cooperative efficiency of unmanned aerial vehicle pilots (operators). In this context, unmanned aerial vehicle simulation technology becomes a key means for pilot training, system verification and human factor assessment. At present, the technical development of the existing unmanned aerial vehicle simulation system mainly focuses on the following layers that firstly, in the aspect of the simulation of an aircraft body, the technology is mature. By establishing a high-precision six-degree-of-freedom (6-DoF) flight dynamics model and combining an aerodynamics, a propulsion system and an automatic control algorithm, the flight attitude, the track and the response characteristic of the unmanned aerial vehicle can be simulated with high fidelity. The system can simulate various flight environments and system faults, such as engine stopping, control surface blocking and the like, and provides an effective platform for verification of a flight control algorithm. Secondly, in terms of task environment and visual simulation, the prior art can construct a high fidelity virtual environment comprising terrain, topography, architecture and meteorological effects by integrating a Geographic Information System (GIS) and a three-dimensional rendering engine. This provides the pilot with an immersive first person perspective (FPV) or third person perspective, assisting in mission planning (e.g., area search, target tracking) and visual perception training. However, despite significant advances in aircraft dynamics and vision simulation, the design philosophy of the prior art generally treats pilots as an ideal, steady-state "operator" model. The existing simulation system is mainly used for recording flight parameters and operation command logs and is used for evaluating external operation performance such as flight trajectory tracking precision, task completion time and the like. However, the operation of an unmanned pilot, particularly in handling high-stress, high-load typical tasks and sudden malfunctions, is closely related to the inherent physiological and psychological states (e.g., psychological load, situational awareness, stress response). The traditional simulation system is difficult to reproduce the influence of the complex cognitive states on the operation, and lacks a platform for synchronously acquiring and relating the flight operation data and the real-time physiological indexes of the pilot. Assessment of pilot performance often relies on post-hoc subjective questionnaires (e.g., NASA-TLX scales), which are hysteresis and subjective and difficult to quantify pilot cognitive load changes in a dynamic task in real-time and objectively. Second, the challenges that pilots face in real tasks are not only flight control, but also comprehensive cognitive processes of information integration, decision making, and emergency handling. Existing fault injection functions are mostly used to verify the reliability of the aircraft itself, rather than being dedicated to evaluating the capability of a person to cope in a loop. Due to the lack of fault setting coupled with task scene depth, capable of precisely controlling trigger time sequence, and synchronous multi-mode data recording capability, the existing system is difficult to scientifically study the complete cognitive chain from fault identification, diagnosis to decision intervention of pilots, and is also difficult to objectively evaluate the influence of different human-computer interaction interface (HCI) designs or operation regulations on pilot performance. In view of the foregoing, a method for constructing realistic tasks and fault scenes, synchronizing, quantifying, and multidimensional monitoring and evaluating operation performance, physiological states and psychological cognition of an unmanned aerial vehicle pilot is needed, so that an indispensable data support and technical means are provided for scientific selection, targeted training, human-computer interface optimization and verification of emergency treatment programs of the unmanned aerial vehicle pilot. Disclosure of Invention The invention aims to provide a pilot state monitoring method and a related device based on unmanned aerial vehicle flight simulation, which are used for solving the technical problem tha