Simulation test system and method of aircraft obstacle avoidance algorithm

Abstract

The application belongs to the technical field of aircraft control, and particularly relates to a simulation test system and method of an aircraft obstacle avoidance algorithm. The system discloses a simulation test system and method of an aircraft obstacle avoidance algorithm, and the system comprises a virtual environment modeling, a dynamics solving, an obstacle avoidance logic processing, a physiological characteristic modeling, a stress parameter mapping, a comprehensive evaluation feedback and a data monitoring device. The system maps kinematic parameters into physiological stress parameters such as gravity overload and motion sickness, and corrects the obstacle avoidance trajectory according to the physiological comfort threshold feedback, so that the obstacle avoidance path meets the human tolerance limit under the premise of safety. The application realizes the deep integration of control engineering and aviation medicine, extends the obstacle avoidance evaluation from physical safety to the physiological tolerance field, improves the scientificity of manned aircraft simulation evaluation, and reduces the risk of personnel disability caused by violent maneuvering in real flight.

Description

Technical Field

[0001] This invention belongs to the field of aircraft control technology, specifically relating to a simulation test system and method for an aircraft obstacle avoidance algorithm. Background Technology

[0002] With the evolution of aerospace technology, the verification of autonomous obstacle avoidance algorithms for aircraft has become a crucial link in ensuring flight safety. Simulation test systems, as the core platform for algorithm evaluation, provide key data support for algorithm iteration and optimization by simulating complex airspace environments and dynamic obstacles. High-performance simulation systems not only require high-precision physics engines but also need to achieve deep integration of aircraft dynamics models and environmental perception logic to ensure the effectiveness and reliability of simulation results under variable weather and complex airspace conditions.

[0003] Simulations of obstacle avoidance algorithms for manned or piloted aircraft not only require the system to have real-time obstacle avoidance capabilities, but also emphasize the profound impact of obstacle avoidance maneuvers on the safety and physiological well-being of passengers. Such simulation systems, by integrating complex motion control logic, aim to simulate the aircraft's maneuverability in extreme avoidance scenarios, ensuring that the flight trajectory meets spatial topological constraints and kinematic performance requirements while complying with basic safety standards for physiological protection in manned aviation.

[0004] Existing obstacle avoidance algorithm simulation models often overemphasize the optimization of geometric paths or the minimization of time costs, lacking quantitative consideration of the physiological limits and tolerance of the human body. Traditional simulation systems, when evaluating maneuvers, cannot establish a dynamic correlation between obstacle avoidance strategies and aviation medical parameters. This results in avoidance paths that, while achieving physical obstacle avoidance at the geometric level, are prone to triggering blackouts or severe motion sickness due to drastic changes in overload parameters. Furthermore, the lack of physiological stress feedback mechanisms makes it difficult to capture the nonlinear mapping relationship between aircraft avoidance maneuvers and human vestibular and visual fatigue, leading to discrepancies between simulation evaluation results and the actual manned flight experience. Summary of the Invention

[0005] The purpose of this invention is to provide a simulation test system for an aircraft obstacle avoidance algorithm, which can solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A simulation test system for an aircraft obstacle avoidance algorithm includes a virtual environment modeling device, an aircraft dynamics calculation device, an obstacle avoidance algorithm logic processing device, a physiological characteristic simulation modeling device, a stress parameter mapping processing device, a comprehensive evaluation feedback adjustment device, and a data monitoring and display device, as follows: The virtual environment modeling device is used to construct a digital simulated airspace containing spatial topology information. Dynamic obstacles with motion vector characteristics and static obstacles with fixed position coordinates are configured in the digital simulated airspace, and a complex meteorological environment including atmospheric turbulence and visibility changes is simulated to provide multi-dimensional external perception input for obstacle avoidance algorithms. The aircraft dynamics calculation device is used to receive control commands in real time and calculate the real-time position, instantaneous velocity, angular velocity and three-axis acceleration components of the aircraft based on the multi-degree-of-freedom dynamic equations. The motion state of the aircraft is fed back to the digital simulation airspace in real time to realize the physical interaction between the aircraft and the virtual environment. The obstacle avoidance algorithm logic processing device is used to run the obstacle avoidance control logic under test. By sensing the obstacle information in the virtual environment modeling device, it generates in real time maneuver commands to change the attitude and trajectory of the aircraft, so as to drive the aircraft to perform avoidance actions and ensure that the flight trajectory meets the non-collision constraints in spatial position. The physiological characteristic simulation modeling device is used to construct a digital virtual human body model in the simulation process to simulate the physiological functions of the personnel in the aircraft during the maneuvering process. The physiological functions cover the vestibular perception characteristics and visual system perception characteristics of the human body, providing a biophysical reference benchmark for assessing the impact of obstacle avoidance actions on personnel. The stress parameter mapping processing device is used to extract the real-time kinematic parameters generated by the aircraft dynamics calculation device and convert them into physiological stress parameters of the digital virtual human body model. The stress parameter mapping processing device can read the virtual human body model parameters in the physiological feature simulation modeling device and complete the accurate mapping of physical motion parameters to physiological stress parameters based on this. The physiological stress parameters include gravity overload tolerance, motion sickness cumulative index and spatial orientation perception offset. By establishing a nonlinear mapping logic between physical motion and physiological response, the physiological load generated by obstacle avoidance actions is quantitatively evaluated. The comprehensive evaluation feedback adjustment device is used to compare the physiological stress parameters with the preset comfort threshold, determine whether the path generated by the current obstacle avoidance algorithm exceeds the physiological tolerance limit of the person, and generate optimization weights to feed back to the obstacle avoidance algorithm logic processing device, so as to make the obstacle avoidance trajectory dynamically corrected in a direction that conforms to physiological suitability while ensuring safe avoidance. The data monitoring and display device is used to record and display in real time the aircraft's avoidance path trajectory, obstacle distance, physiological stress parameter change curves, and comprehensive evaluation score. It also presents the simulation evaluation performance of the obstacle avoidance algorithm in a human-machine closed-loop scenario through a graphical interface.

[0007] Preferably, the virtual environment modeling device is equipped with a meteorological interference generation module, which is used to introduce randomly generated disturbance vectors into the digital simulation airspace. The disturbance vectors cause displacement of the aircraft's position coordinates according to a preset airflow intensity distribution law, simulating the interference of sudden airflow on obstacle avoidance accuracy and personnel stability in the real flight environment.

[0008] Furthermore, when performing state calculations, the aircraft dynamics calculation device not only considers the displacement of the aircraft's center of mass, but also comprehensively considers the moment of inertia of the aircraft about its own center of mass. This ensures that the angular acceleration generated during high-G evasive maneuvers reflects the true physical laws, providing accurate raw motion data for subsequent mapping of physiological parameters.

[0009] Furthermore, the physiological characteristic simulation modeling device includes a vestibular model. The vestibular model converts the linear acceleration and angular acceleration of the aircraft into acceleration perception signals of the virtual human body by simulating the physical perception mechanism of the semicircular canals and otoliths in the human inner ear, and determines the degree of deviation between the perception signal and the real physical motion signal in order to predict the probability of spatial orientation disorder.

[0010] Furthermore, when calculating the gravity overload tolerance, the stress parameter mapping processing device uses the sum of square roots of the squares of the three-axis acceleration components extracted in real time to obtain the resultant acceleration value, and performs integration processing in combination with the duration dimension. When the value after integration processing exceeds the preset gravity tolerance threshold and the duration exceeds the predetermined time length, the system automatically determines that the current maneuver will cause the personnel inside the aircraft to have a risk of physiological compensatory disorder.

[0011] Furthermore, when assessing the motion sickness cumulative index, the stress parameter mapping processing device establishes a relationship function based on the low-frequency oscillation frequency and duration, extracts the reciprocating oscillation motion generated by the aircraft within a certain frequency range during obstacle avoidance, and obtains a quantitative index reflecting the intensity of motion sickness by performing time-domain cumulative calculation on the vibration amplitude within this frequency range. When this index exceeds the preset comfort warning line, the intervention logic of the comprehensive evaluation feedback adjustment device is triggered.

[0012] Furthermore, the comprehensive evaluation feedback adjustment device establishes a set of multi-objective optimization evaluation criteria. The evaluation criteria assign different weight coefficients to geometric obstacle avoidance safety, time optimality, and physiological comfort. By calculating the comprehensive cost value, it guides the obstacle avoidance algorithm to search for the trajectory scheme with the minimum comprehensive cost value among multiple feasible paths.

[0013] Preferably, the obstacle avoidance algorithm logic processing device has dynamic learning capability. Based on the weight correction value provided in real time by the comprehensive evaluation feedback adjustment device, it automatically adjusts the constraint limits of turning radius, climb rate and descent rate during the path planning process to ensure that the smoothness of the maneuver is improved in the subsequent avoidance maneuver and avoids the generation of instantaneous overload pulses.

[0014] Furthermore, the data monitoring and display device supports multi-channel synchronous playback, which can frame-align the first-view image during obstacle avoidance with the dynamic curve of physiological stress parameters, so that technicians can identify the specific obstacle avoidance stage that causes physiological discomfort and the corresponding type of maneuver.

[0015] Furthermore, the virtual environment modeling device is also equipped with random obstacle generation logic, which can release interference objects with uncertain motion trajectories in real time on the expected trajectory of the aircraft according to a preset collision probability function during the simulation process, so as to test the robustness of the obstacle avoidance algorithm and physiological feedback system in dealing with extreme emergencies.

[0016] Furthermore, the stress parameter mapping processing device is also equipped with visual fatigue assessment logic. This logic calculates the visual cognitive load of the virtual human body by extracting the refresh rate and contrast changes of the virtual display interface in the simulation environment and the visual sweep speed generated by the target movement. It is used as a supplementary component of the physiological stress parameters to evaluate the human-computer interaction friendliness during obstacle avoidance.

[0017] A simulation test method for an aircraft obstacle avoidance algorithm is provided, which uses the simulation test system for the aircraft obstacle avoidance algorithm as described in any one of the claims to conduct simulation tests on the aircraft obstacle avoidance algorithm.

[0018] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention breaks through the limitations of traditional obstacle avoidance algorithm simulation, which only focuses on physical space safety and efficiency. By introducing a physiological feature simulation modeling device and a stress parameter mapping processing device, it achieves a deep integration of control engineering and aviation medicine, so that the simulation evaluation of obstacle avoidance algorithm is not limited to avoiding collisions, but extends to a deeper field that conforms to human physiological tolerance.

[0019] 2. The human-machine closed-loop adaptive feedback mechanism constructed in this invention can quantify abstract physiological reactions (such as blackout risk and motion sickness index) into mathematical feedback that can be used for algorithm optimization. This enables the obstacle avoidance algorithm to have human-like flight characteristics in the design stage, reducing the risk of pilot incapacitation or extreme passenger discomfort caused by excessive algorithm maneuvers in real manned flight missions, and improving the practicality of the system.

[0020] 3. Through the joint simulation of a high-precision physics engine and a physiological model, this invention can accurately simulate the subtle changes in physical parameters of an aircraft under extreme maneuvering scenarios and their nonlinear effects on personnel, providing a more scientific, comprehensive and ergonomically compliant evaluation method for the safety verification of manned aircraft's assisted driving system, autonomous navigation system, and unmanned aerial vehicle remote control system.

[0021] 4. The system architecture of this invention is flexible and scalable. By adjusting the thresholds and parameter configurations in the physiological characteristic model, it can simulate the stress response of people with different physical conditions, ages, and training backgrounds during obstacle avoidance, enriching the coverage of simulation experiments and providing data support for the development of customized aircraft control logic.

[0022] 5. The data monitoring and display device enables the synchronous visualization of physiological and flight parameters, which not only improves the transparency of the simulation process, but also provides technicians with an intuitive and quantitative auxiliary decision-making tool to seek the optimal balance between safety, efficiency and comfort through multi-objective comprehensive cost-benefit assessment. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the overall technical solution architecture of the present invention; Figure 2 This is a schematic diagram of the core principle framework of mapping physical kinematic parameters to digital virtual human physiological stress parameters in this invention; Figure 3 This is a logical flowchart of the digital simulation airspace construction and complex meteorological environment simulation in this invention; Figure 4 This is a schematic diagram of the multi-level interaction relationship and data flow of obstacle avoidance algorithm logic processing, physiological feature simulation and comprehensive evaluation feedback in this invention. Detailed Implementation

[0024] Example 1: Please refer to the appendix Figure 1 To be continued Figure 4 A simulation test system for an obstacle avoidance algorithm for aircraft includes a virtual environment modeling device, an aircraft dynamics calculation device, an obstacle avoidance algorithm logic processing device, a physiological characteristic simulation modeling device, a stress parameter mapping processing device, a comprehensive evaluation feedback adjustment device, and a data monitoring and display device.

[0025] The virtual environment modeling device, serving as the spatial foundation of the entire simulation system, is configured to construct a digital simulation airspace containing spatial topology information. At the hardware implementation level, the virtual environment modeling device is deployed in a high-performance graphics workstation, generating complex 3D scenes with geographic coordinate information through the collaboration of a 3D rendering engine and a spatial database. The virtual environment modeling device integrates a spatial topology construction module, which defines the boundaries of the airspace, no-fly zones, and airway networks within the virtual coordinate system. The virtual environment modeling device also includes a dynamic obstacle configuration unit and a static obstacle configuration unit. The dynamic obstacle configuration unit generates obstacles with motion vector characteristics within the digital simulation airspace, such as other aircraft, birds, or floating objects, and assigns them real-time velocity vectors, heading angles, and trajectory equations. The static obstacle configuration unit configures obstacles with fixed position coordinates, such as mountains, buildings, high-voltage power towers, and trees, simulating their physical outlines using a high-precision terrain mesh model.

[0026] Furthermore, the virtual environment modeling device also includes a meteorological environment simulation unit. This unit is configured to simulate a complex meteorological environment including atmospheric turbulence, visibility variations, and crosswind interference. The unit incorporates a meteorological disturbance generation module, used to introduce randomly generated disturbance vectors into the digital simulation airspace. These disturbance vectors, based on a preset airflow intensity distribution pattern (such as a parameterized transformation of the DeWitt turbulence model or the Karman turbulence model), cause displacement shifts in the aircraft's position coordinates. In describing the disturbance logic, the system calculates the product of the atmospheric disturbance intensity coefficient and random Gaussian white noise to obtain torque disturbance components along the three axes. This simulates the interference of sudden airflow on obstacle avoidance accuracy and passenger stability in a real flight environment, providing the obstacle avoidance algorithm with multi-dimensional external sensory inputs that combine high randomness and determinism.

[0027] The aircraft dynamics calculation device is connected to the virtual environment modeling device via a real-time data bus and is used to receive control commands from the obstacle avoidance algorithm logic processing device. Based on multi-degree-of-freedom dynamic equations, the aircraft dynamics calculation device calculates the aircraft's motion state in real time. Logically, the device considers not only the displacement of the aircraft's center of mass in three-dimensional space (i.e., changes in longitude, latitude, and altitude), but also the rotational inertia of the aircraft about its own center of mass, encompassing the rotational dynamics of roll, pitch, and yaw.

[0028] The aircraft dynamics calculation device includes a mass attribute configuration unit for defining the aircraft's mass, center of mass position, and inertia tensor matrix. During state calculation, the calculation kernel is configured to solve the nonlinear dynamic differential equations using numerical integration methods (such as the fourth-order Runge-Kutta method). Specifically, the device calculates the aircraft's three-axis acceleration components in real time: axial acceleration, lateral acceleration, and normal acceleration; it also calculates the three-axis angular acceleration to ensure that the generated physical data reflects real physical laws during high-G evasive maneuvers (such as sharp climbs or steep turns). This precise raw motion data, including instantaneous velocity, angular velocity, and the resultant G-force generated by the maneuver, is fed back in real time to the digital simulation airspace and simultaneously transmitted to the stress parameter mapping processing device, enabling physical interaction between the aircraft and the virtual environment.

[0029] The obstacle avoidance algorithm logic processing device is used to run the obstacle avoidance control logic under test. As the main carrier of the algorithm, this device senses obstacle information in the virtual environment modeling device and generates maneuver commands in real time to change the aircraft's attitude and trajectory. The obstacle avoidance algorithm logic processing device integrates a perception interface module, a path planning module, and a command generation module. The perception interface module extracts the coordinates and motion vectors of obstacles within a preset perception radius from the virtual environment modeling device; the path planning module calculates the optimal avoidance path based on specific algorithm logic (such as the artificial potential field method, velocity obstacle method, or deep reinforcement learning model), ensuring that the flight trajectory meets the non-collision constraints in spatial position, that is, the Euclidean distance between the aircraft and any obstacle is always greater than the preset safe distance.

[0030] Furthermore, the obstacle avoidance algorithm logic processing device possesses dynamic learning capabilities and parameter adaptive adjustment functions. This device is configured to automatically adjust the constraint limits for turning radius, climb rate, and descent rate during path planning based on the weight correction values ​​provided in real time by the comprehensive evaluation feedback adjustment device. For example, when the comprehensive evaluation feedback adjustment device determines that the current maneuver overload exceeds a preset range, the obstacle avoidance algorithm logic processing device will automatically increase the curvature radius of the planned path, sacrificing some path length to achieve a smoother maneuver and avoid generating sudden, drastic overload pulses.

[0031] The physiological characteristic simulation modeling device is one of the core components of this system for achieving closed-loop human-machine simulation. This device is used to construct a digital virtual human body model during the simulation process, simulating the physiological functions of personnel inside the aircraft during aircraft maneuvers. These physiological functions encompass the vestibular perception characteristics and visual system perception characteristics.

[0032] The physiological characteristic simulation modeling device includes a vestibular model submodule. This submodule establishes biomechanical response logic by simulating the physical perception mechanism of the semicircular canals and otolith organs in the human inner ear. The vestibular model maps the linear acceleration input of the aircraft to the displacement perception of the otolith organs and the angular acceleration input to the flow perception of lymph in the semicircular canals. This model can determine the degree of deviation between the perceived signal (i.e., the motion subjectively felt by the virtual human body) and the real physical motion signal. When the deviation exceeds a threshold, the physiological characteristic simulation modeling device predicts the probability of spatial orientation disorder. The device also includes a visual system perception model to simulate changes in visual cognitive load caused by narrowed field of vision, blurred vision, or information overload under high-dynamic maneuvers.

[0033] The stress parameter mapping processing device receives real-time kinematic parameters generated by the aircraft dynamics calculation device through a high-speed data path and converts them into physiological stress parameters of the digital virtual human model. The stress parameter mapping processing device can read the virtual human model parameters in the physiological feature simulation modeling device and, based on this, complete the accurate mapping of physical motion parameters to physiological stress parameters; the physiological stress parameters include gravity overload tolerance, motion sickness cumulative index, and spatial orientation perception offset.

[0034] When calculating gravity overload tolerance, the stress parameter mapping processing device is configured to: utilize the real-time extracted triaxial acceleration components, calculate the arithmetic square root of the sum of the squares of each component, and obtain the resultant acceleration value; weight the resultant acceleration value with the projection of the human body along the longitudinal axis to obtain the effective overload value acting on the human body. The system further performs integral processing in conjunction with the duration dimension to calculate the overload time integral curve. When the value after the integral processing exceeds a preset gravity tolerance threshold (such as 3 times the gravitational acceleration for ordinary personnel for more than 5 seconds, or a higher threshold for professional pilots), and the duration exceeds a predetermined time length, the system automatically determines that the current maneuver will cause physiological compensatory disorder risks to the occupants, such as graying, blackout, or loss of consciousness.

[0035] When assessing the cumulative motion sickness index, the stress parameter mapping processing device establishes a function based on the relationship between low-frequency oscillation frequency and duration. The system is configured to extract the reciprocating oscillation motion (especially low-frequency vibrations between 0.1 Hz and 0.5 Hz) generated by the aircraft during obstacle avoidance within a certain frequency range. By performing time-domain cumulative calculations on the vibration amplitude within this frequency range, a quantitative index reflecting the intensity of motion sickness (such as the MSDV value) is obtained. When this index exceeds a preset comfort threshold, subsequent intervention logic is triggered.

[0036] The stress parameter mapping processing device is also equipped with visual fatigue assessment logic. This logic calculates the visual cognitive load of the virtual human body by extracting the refresh rate and contrast changes of the virtual display interface in the simulation environment, as well as the visual sweep speed generated by the target movement (i.e., the speed at which the image slides on the retina). This load value serves as a supplementary component of the physiological stress parameters and is used to comprehensively evaluate the human-computer interaction friendliness during obstacle avoidance.

[0037] The comprehensive evaluation feedback adjustment device is used to compare the physiological stress parameters with a preset comfort threshold to determine whether the path generated by the current obstacle avoidance algorithm exceeds the physiological tolerance limit of the user. This device establishes a multi-objective optimization evaluation criterion, which assigns different weight coefficients to geometric obstacle avoidance safety, time optimality, and physiological comfort.

[0038] In the specific feedback adjustment logic, the comprehensive evaluation feedback adjustment device calculates the comprehensive cost value. This comprehensive cost value is equal to the weighted sum of the safety penalty, time cost, and physiological load. The safety penalty is inversely proportional to the minimum distance between the aircraft and the obstacle; the time cost is directly proportional to the increased flight distance due to the avoidance maneuver; and the physiological load is a weighted sum of the aforementioned gravity overload, motion sickness index, and visual fatigue. The system compares the comprehensive cost values ​​of different potential avoidance paths in real time, generates optimized weights, and feeds them back to the obstacle avoidance algorithm logic processing device. This prompts the obstacle avoidance trajectory to be dynamically corrected in a direction that conforms to physiological suitability while ensuring safe avoidance. This mechanism ensures that human-centered constraints are incorporated into the design phase of the obstacle avoidance algorithm.

[0039] The data monitoring and display device, deployed in a multi-screen display system or virtual reality headset, is used to record and display key data of the simulation process in real time and in multiple dimensions. The device can simultaneously present the aircraft's avoidance path trajectory, obstacle spacing, changes in physiological stress parameters, and comprehensive evaluation scores.

[0040] Preferably, the data monitoring and display device supports multi-channel synchronous playback. This function is configured to display the first-person view during obstacle avoidance (i.e., the external view seen by the virtual driver) frame-aligned with dynamic curves of physiological stress parameters (such as acceleration curves, simulated heart rate values, motion sickness scores, etc.). This synchronous display mechanism allows technicians to accurately identify the specific obstacle avoidance stage that causes physiological discomfort. For example, technicians can identify that during a rapid roll maneuver at a specific angle, a sudden change in angular acceleration triggers a spatial orientation obstacle alarm in the vestibular model, allowing for targeted optimization of the attitude control gain in the algorithm.

[0041] To further enhance the comprehensiveness of the simulation, the virtual environment modeling device is also equipped with random obstacle generation logic. This logic can, during the simulation, deploy interfering objects with uncertain trajectories in real time along the aircraft's expected trajectory, based on a preset collision probability function. This not only tests the geometric robustness of the obstacle avoidance algorithm under extreme emergency conditions but also tests the effectiveness of the physiological feedback system in assessing the cumulative physiological load during continuous high-frequency emergency maneuvers.

[0042] During system operation, data flow between various devices forms a closed loop. The virtual environment modeling device sends obstacle space data to the obstacle avoidance algorithm logic processing device; the obstacle avoidance algorithm logic processing device calculates control commands and sends them to the aircraft dynamics calculation device; the motion parameters calculated by the aircraft dynamics calculation device are fed back to the virtual environment for position updates and sent to the stress parameter mapping processing device; the stress parameter mapping processing device, in conjunction with the physiological characteristic simulation modeling device, generates physiological load data; and the comprehensive evaluation feedback adjustment device corrects the obstacle avoidance algorithm logic in real time based on the physiological load data.

[0043] This human-machine closed-loop simulation architecture changes the traditional obstacle avoidance algorithm's single evaluation dimension of only pursuing physical non-collision. By introducing an aviation medical model into the control loop, it achieves deep coupling between obstacle avoidance performance and personnel safety.

[0044] Example 2: As a further supplement and architectural variant of Example 1, Example 2 describes a simulation test system for an aircraft obstacle avoidance algorithm based on a distributed computing architecture. In this example, the various functional devices of the system are distributed on different hardware nodes and interact with each other at high speed via deterministic industrial Ethernet to meet the real-time requirements of physiological characteristic simulation at high sampling frequencies.

[0045] In this embodiment, the virtual environment modeling device is deployed on a dedicated terrain server. This server is equipped with multiple high-performance graphics processing units (GPUs) to achieve real-time rendering of large-scale complex scenes using hardware acceleration technology. The terrain server stores a high-precision global digital elevation model (DEM) and can dynamically load local refined meshes based on the aircraft's current latitude and longitude coordinates. To improve simulation accuracy, the meteorological disturbance generation module in the virtual environment modeling device employs distributed fluid dynamics computing nodes. By pre-calculating flow field data and combining it with real-time disturbance interpolation, it provides unsteady atmospheric environmental parameters for dynamic solution.

[0046] The aircraft dynamics calculation device is deployed in a real-time simulator. This simulator uses an operating system with hard real-time characteristics, ensuring that the iteration period of the dynamic equations remains stable within 1 millisecond. To simulate different types of manned aircraft, the aircraft dynamics calculation device is configured to support switching between Model-in-the-Loop (MIL) and Hardware-in-the-Loop (HIL) modes. In Hardware-in-the-Loop mode, the actual flight control computer can access the system via a CAN bus or ARINC429 bus, replacing some of the functions of the obstacle avoidance algorithm logic processing device, and verifying the impact of the dynamic response characteristics of the actual control hardware when executing obstacle avoidance commands on the physiological state of personnel.

[0047] The physiological characteristic simulation modeling device is further enhanced in this embodiment. Its digital virtual human body model not only covers vestibular and visual senses, but also extends to the simulation of the circulatory and musculoskeletal systems. The device is configured as an independent biomechanical calculation unit, internally storing sets of physiological characteristic parameters for individuals with different physical conditions (such as different ages, body mass indexes, and training levels).

[0048] The physiological characteristic simulation modeling device includes a cardiovascular system simulation module. This module calculates the changes in blood pressure at the oculobronchial level of the virtual human body based on the longitudinal overload data output by the stress parameter mapping processing device. When the calculated oculobronchial arterial pressure falls below a preset critical perfusion pressure, the system determines that physiological disability (such as blackout) has occurred. This in-depth physiological simulation allows the system to assess the impact of the overload gradient (i.e., the rate of change of G value over time) generated by the obstacle avoidance algorithm on the safety of personnel inside the aircraft.

[0049] In this embodiment, the stress parameter mapping processing device employs a multi-level mapping architecture. The first-level mapping is kinematic parameter preprocessing, used to filter high-frequency vibration signals from the aircraft dynamics calculation device, removing high-frequency structural noise that has little physiological impact. The second-level mapping is physiological excitation conversion, converting the filtered smooth acceleration and angular acceleration data into vestibular sensory stimulation.

[0050] For assessing the cumulative motion sickness index, the stress parameter mapping processing device in this embodiment employs a computational model based on sensory conflict theory. This model is configured to calculate the vector difference between the motion sensation predicted by the visual system model in the physiological feature simulation modeling device and the motion sensation predicted by the vestibular model. This difference represents information inconsistency in human brain perception. The stress parameter mapping processing device performs a weighted integral on this difference, and when the integral value reaches a preset vomiting incidence threshold, it issues a comfort warning to the comprehensive evaluation feedback adjustment device.

[0051] In this embodiment, the comprehensive evaluation feedback adjustment device is implemented as a decision support server. This server runs a multi-objective search algorithm based on the Pareto optimality principle. When the obstacle avoidance algorithm logic processing device submits multiple candidate obstacle avoidance paths, the decision support server calculates the score for each path in terms of collision probability, task time, peak overload, and cumulative motion sickness value.

[0052] The comprehensive evaluation feedback adjustment device is configured to select the optimal path by calculating a comprehensive cost function. The comprehensive cost function is defined as the sum of the products of each evaluation index and its corresponding weight coefficient. In special mission scenarios, such as medical evacuation flights or transporting valuables, users can adjust the weight coefficient of physiological comfort in real time through the interactive interface of the data monitoring and display device. The comprehensive evaluation feedback adjustment device will guide the obstacle avoidance algorithm to adopt a gentler, less maneuverable avoidance logic through feedback commands, even if this may result in a longer obstacle avoidance flight path.

[0053] In this embodiment, the data monitoring and display device employs distributed display technology. The main control console display system is responsible for presenting a three-dimensional global airspace situation map, demonstrating the relative spatial relationships between the aircraft and various dynamic and static obstacles. The side-wing displays are specifically designed to display the physiological monitoring panel. This panel simulates the interface of a real aviation medical monitor, displaying the virtual human's real-time heart rate, blood oxygen saturation trends, and vestibular conflict index.

[0054] The data monitoring and display device in this embodiment also includes an automatic report generation module. After the simulation test, this module is configured to automatically analyze the physiological data records throughout the obstacle avoidance process. The report automatically marks all timestamps that violate physiological comfort criteria and extracts the corresponding maneuver characteristics (such as maximum pitch rate, peak combined overload, etc.). Through cluster analysis of these characteristics, the system can identify the logical design defects of the current obstacle avoidance algorithm.

[0055] This embodiment also introduces a physiological feedback mode for UAV remote operators. In this mode, the physiological feature simulation modeling device simulates the cognitive load of the ground operator when receiving delayed video streams and control feedback. The stress parameter mapping processing device calculates the inconsistency between visual and control perception caused by link latency, and evaluates the operability of the obstacle avoidance algorithm in a remote control environment. This allows the system to not only assess the comfort of the personnel inside the drone, but also the impact of obstacle avoidance logic on the psychological stress of the operator, achieving a more generalized human-machine closed-loop simulation.

[0056] Through this distributed hardware deployment and deep biomechanical extension, the system described in Example 2 can provide a more refined and realistic verification platform for the development of obstacle avoidance algorithms for high-performance aircraft.

[0057] Example 3: Based on Examples 1 and 2, this example further describes in detail the extended application and module configuration of the simulation test system of the aircraft obstacle avoidance algorithm in dealing with multi-aircraft cooperative obstacle avoidance scenarios.

[0058] In this embodiment, the virtual environment modeling device is configured to support multi-agent cooperative simulation. This means that the digital simulation airspace contains not only a single test aircraft, but also multiple cooperative aircraft with independent dynamic characteristics and obstacle avoidance logic. The topology building module inside the virtual environment modeling device is responsible for maintaining the spatial proximity matrix between all aircraft in real time, providing local situational awareness data for each aircraft.

[0059] In Embodiment 3, the aircraft dynamics calculation device is instantiated as multiple parallel calculation threads. Each thread is responsible for the dynamics simulation of a specific aircraft. To reflect the differences in physiological tolerance among different aircraft types, the system allows different digital virtual human models to be associated with each aircraft. For example, the navigator aircraft can be associated with a high-tolerance professional pilot model, while the following civilian transport aircraft can be associated with an ordinary passenger model.

[0060] The physiological characteristic simulation modeling device adds a group collaboration stress assessment submodule for multi-machine collaborative scenarios. This submodule is configured to simulate the psychological stress response caused by excessively close proximity of multiple machines during dense formation obstacle avoidance. This stress response is converted into heart rate rise rate and decision-making reaction delay time through a stress parameter mapping and processing device.

[0061] In this embodiment, the stress parameter mapping processing device, in addition to calculating the G-force overload of a single aircraft, also calculates the consistency physiological indicators generated by formation maneuvers. The device is configured to calculate the average motion sickness index of all members within the formation and the variance of that index. If the variance is too large, it indicates that the current obstacle avoidance strategy has led to an imbalance in the physiological load of the personnel within the formation (e.g., aircraft located at the edge of the formation need to perform more intense maneuvers).

[0062] The comprehensive evaluation feedback adjustment device employs a distributed coordination game logic in multi-aircraft scenarios. The device is configured not only to minimize the cost function of a single aircraft, but also to minimize the total physiological load of the entire formation. When an aircraft's obstacle avoidance path would cause its personnel to experience blackout risk, the comprehensive evaluation feedback adjustment device sends a coordination request to surrounding aircraft, prompting other aircraft to voluntarily give way and allow the restricted aircraft to adopt a smoother avoidance trajectory.

[0063] Specifically, during execution, the comprehensive evaluation feedback adjustment device calculates the collaborative evaluation index. The formula for this index is: ; in, As a collaborative evaluation indicator, For the first Safety score of the aircraft No. The physiological comfort score of the aircraft. For the product of security scores, For the first The weighting coefficients for the physiological comfort score of the aircraft should meet the following requirements. , 0; Through this nonlinear coupling method, the system can force the obstacle avoidance algorithm to prioritize protecting aircraft that are on the verge of physiological limits in extreme scenarios.

[0064] In this embodiment, the data monitoring and display device adds a formation situation overview interface. This interface uses a heat map to display the physiological risk level of different areas within the airspace. For example, when a certain airspace has turbulent airflow and dense obstacles, that area is marked in red on the display interface, indicating that performing obstacle avoidance maneuvers in this area will generate extremely high physiological stress for personnel.

[0065] The data monitoring and display device in this embodiment supports virtual reality (VR) collaborative analysis. Multiple technicians can wear VR headsets to enter the simulation space and observe the formation obstacle avoidance process from different perspectives. The system displays the data output by the stress parameter mapping and processing device in real time as a semi-transparent floating window above each aircraft, achieving a coupled presentation of physical motion and physiological stress.

[0066] In multi-aircraft mode, the obstacle avoidance algorithm logic processing device extends its dynamic learning capability to the group level. This device is configured to learn how to distribute maneuver intensity among formation members through a shared experience pool. For example, during long-endurance missions, the algorithm automatically rotates the positions of aircraft performing high-G obstacle avoidance maneuvers to prevent premature fatigue in a single crew due to continuous high-G maneuvers or frequent motion sickness.

[0067] Through this multi-aircraft collaborative expansion, the simulation system provided by this invention can provide a valuable means of verifying obstacle avoidance strategies for future urban air traffic (UAM) and complex formation flight missions, ensuring that the physiological safety and comfort of each passenger can be quantitatively guaranteed in complex group flight environments.

[0068] In all the above embodiments, the logical relationships and physical functions of each functional device and its sub-module of the system are described in text form, without relying on any specific mathematical formulas. All comparison, accumulation, mapping, and weight adjustment logic involved has been translated into detailed Chinese engineering language descriptions. For example, the calculation process involving squares and square roots is described as calculating the arithmetic square root of the sum obtained by squaring each component value; the process involving threshold determination is described as determining whether a specific physiological stress value exceeds a preset safety warning limit.

[0069] The system architectures described in each embodiment can be implemented based on general-purpose computers, industrial controllers, embedded processing boards, and dedicated sensor simulation interfaces. Data interaction between devices follows standard communication protocols, such as Ethernet, bus communication, or shared memory mechanisms.

[0070] In this invention, obstacle avoidance algorithm simulation is no longer an isolated geometric planning process, but an organic whole that tightly integrates aircraft physical motion, airspace meteorological environment, and human biological response. By introducing a physiological characteristic simulation modeling device and a stress parameter mapping processing device, the system can capture physiological insecuritys that traditional simulation methods cannot detect, providing a rigorous and scientific evaluation system for manned aircraft algorithm development. This deeply integrated architecture embodies an advanced form of control engineering and biomedical engineering in the aerospace field, and has significant practical engineering implications for improving the autonomous flight safety and occupant suitability of aircraft.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications can still be made to the specific hardware implementation of the present invention or equivalent substitutions can be made to some technical features. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of the claims of the present invention.

[0072] Those skilled in the art will understand that the above-described division of devices and modules is merely a schematic representation of logical functions. In actual product development, they can be integrated or further separated according to hardware performance and engineering requirements. For example, the function of the stress parameter mapping and processing device can be integrated into the physiological characteristic simulation modeling device, or the comprehensive evaluation feedback adjustment device can be deployed on a cloud server to support larger-scale simulation calculations. These variations based on the core concept of this invention all fall within the protection scope of this patent.

[0073] In practical applications, the system described in this invention can be parametrically customized for different types of aircraft. For example, for high-maneuverability fighter jets, the gravity tolerance threshold in the physiological characteristic simulation modeling device can be set higher, with a focus on the auxiliary assessment of anti-G-force maneuvering (AGSM); while for civilian passenger drones, the system increases the weight of the motion sickness accumulation index in the comprehensive evaluation feedback adjustment device to ensure a smooth experience for passengers. This flexibility ensures that this invention can be widely applied in various fields such as general aviation, special flights, and future unmanned air transport.

[0074] In summary, this invention solves the problem of disconnect between obstacle avoidance algorithm simulation and human physiological perception in existing technologies by constructing a complete simulation closed loop that includes physiological feedback, providing strong technical support for ergonomics research and algorithm safety verification in the aerospace field.

[0075] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A simulation test system for an obstacle avoidance algorithm for aircraft, characterized in that, It includes a virtual environment modeling device, an aircraft dynamics calculation device, an obstacle avoidance algorithm logic processing device, a physiological characteristic simulation modeling device, a stress parameter mapping processing device, a comprehensive evaluation feedback adjustment device, and a data monitoring and display device, among which: The virtual environment modeling device is used to construct a digital simulated airspace containing spatial topology information, and to configure dynamic obstacles with motion vector characteristics, static obstacles with fixed position coordinates, and a simulated meteorological environment containing atmospheric turbulence changes within the digital simulated airspace. The aircraft dynamics calculation device is used to calculate the motion state of the aircraft in real time based on multi-degree-of-freedom dynamic equations, and to feed back the motion state to the digital simulation airspace in real time. The aircraft dynamics calculation device is connected to the virtual environment modeling device via a real-time data bus. The aircraft dynamics calculation device is equipped with a mass attribute configuration unit, which is used to define the mass, center of mass position and inertia tensor matrix of the aircraft. During the state calculation, the calculation kernel of the aircraft dynamics calculation device is configured to solve the nonlinear dynamic differential equation system by numerical integration. In the calculation process, it not only calculates the displacement changes of the aircraft's center of mass in longitude, latitude and altitude in three-dimensional space, but also comprehensively considers the rotational inertia of the aircraft about its own center of mass in three axes: roll, pitch and yaw. The motion state parameters output in real time by the aircraft dynamics calculation device include the aircraft's real-time position, instantaneous velocity, angular velocity, three-axis angular acceleration, and three-axis acceleration components including axial acceleration, lateral acceleration, and normal acceleration, ensuring that the physical data generated during high-G avoidance maneuvers reflect the physical laws of the aircraft. The obstacle avoidance algorithm logic processing device is used to run the obstacle avoidance control logic under test. By sensing the obstacle information in the virtual environment modeling device, it generates maneuver commands in real time to change the attitude and trajectory of the aircraft, and drives the aircraft to perform avoidance actions. The physiological feature simulation modeling device is used to construct a digital virtual human body model in the simulation process to simulate the physiological functions of the personnel inside the aircraft during the aircraft maneuver. The physiological functions include the vestibular perception characteristics and the visual system perception characteristics of the human body. The stress parameter mapping processing device is used to convert the kinematic parameters generated by the aircraft dynamics solving device into the physiological stress parameters of the digital virtual human body model; the stress parameter mapping processing device can read the virtual human body model parameters in the physiological feature simulation modeling device, and based on this, complete the accurate mapping of physical motion parameters to physiological stress parameters. The physiological characteristic simulation modeling device includes a vestibular model submodule and a visual system perception model. The vestibular model submodule establishes biomechanical response logic by simulating the physical perception mechanism of the semicircular canals and otoliths in the human inner ear, mapping the linear acceleration input of the aircraft to the displacement perception of the otoliths, and mapping the angular acceleration input to the flow perception of lymph in the semicircular canals. The vestibular model submodule is used to determine the degree of deviation between the motion perception signal subjectively felt by the digital virtual human body and the real physical motion signal, and when the degree of deviation exceeds a threshold, it predicts the probability of spatial orientation disorder. The visual system perception model is used to simulate changes in visual cognitive load caused by narrowed field of view, blurred vision, or information overload in highly dynamic maneuvering scenarios. The stress parameter mapping and processing device receives real-time kinematic parameters through a high-speed data path. The physiological stress parameters include gravity overload tolerance, motion sickness cumulative index, and spatial orientation perception offset. When calculating the gravity overload tolerance, the stress parameter mapping processing device is configured to: firstly, using the real-time extracted triaxial acceleration components, calculate the arithmetic square root of the sum of the squared values ​​of each component to obtain the resultant acceleration value. The combined acceleration value is weighted and correlated with the projection of the human body along the longitudinal axis to obtain the effective overload value acting on the human body. The effective overload value is integrated by combining the duration dimension to calculate the overload time integral curve; When the value corresponding to the overload time integral curve exceeds the preset gravity tolerance threshold and the duration exceeds the predetermined time length, the system automatically determines that the current maneuver causes the digital virtual human body model to have a risk of physiological compensation disorder. The comprehensive evaluation feedback adjustment device is used to compare the physiological stress parameters with the preset comfort threshold and generate optimized weights to feed back to the obstacle avoidance algorithm logic processing device, so as to make the obstacle avoidance trajectory dynamically corrected in a direction that conforms to physiological suitability. The data monitoring and display device is used to record and display in real time the aircraft's avoidance path trajectory, obstacle spacing, physiological stress parameter change curves, and comprehensive evaluation score.

2. The simulation test system for the aircraft obstacle avoidance algorithm according to claim 1, characterized in that, The virtual environment modeling device is deployed in a graphics workstation and generates a 3D scene with geographic coordinate information through the collaboration of a 3D rendering engine and a spatial database. The virtual environment modeling device integrates a spatial topology construction module, a dynamic obstacle configuration unit, a static obstacle configuration unit, and a meteorological environment simulation unit. The spatial topology construction module is used to define the boundary range of the airspace, no-fly zones, and airway networks in a virtual coordinate system; The dynamic obstacle configuration unit is used to assign real-time velocity vectors, heading angles, and motion trajectory equations to dynamic obstacles; The static obstacle configuration unit uses a high-precision terrain mesh model to simulate the physical outlines of mountains, buildings, high-voltage power towers, and trees; The meteorological environment simulation unit is equipped with a meteorological interference generation module, which is used to introduce randomly generated disturbance vectors into the digital simulation airspace. The disturbance vectors are obtained by calculating the product of atmospheric disturbance intensity coefficient and random Gaussian white noise to obtain torque interference components in the three-axis direction, and the displacement of the aircraft's position coordinates is generated according to the preset airflow intensity distribution law.

3. The simulation test system for the aircraft obstacle avoidance algorithm according to claim 1, characterized in that, The obstacle avoidance algorithm logic processing device integrates a perception interface module, a path planning module, and an instruction generation module. The perception interface module is used to extract the coordinates and motion vectors of obstacles within a preset perception radius from the virtual environment modeling device. The path planning module calculates the avoidance path based on artificial potential field logic, speed obstacle logic, or deep reinforcement learning model, ensuring that the Euclidean distance between the aircraft and any obstacle is always greater than the preset safe interval distance. The obstacle avoidance algorithm logic processing device has dynamic learning capability and parameter adaptive adjustment function. It is configured to automatically adjust the constraint limits of turning radius, climbing rate and descent rate in the path planning process according to the weight correction value provided in real time by the comprehensive evaluation feedback adjustment device. When the comprehensive evaluation feedback adjustment device determines that the current maneuver overload exceeds the preset range, the obstacle avoidance algorithm logic processing device increases the radius of curvature of the planned path, and improves the smoothness of the maneuver process by sacrificing part of the path length, so as to avoid generating instantaneous and drastic overload pulses.

4. The simulation test system for the aircraft obstacle avoidance algorithm according to claim 1, characterized in that, When assessing the motion sickness cumulative index, the stress parameter mapping processing device establishes a relationship function based on the low-frequency oscillation frequency and duration. The stress parameter mapping processing device is configured to extract the reciprocating oscillation motion generated by the aircraft in the frequency range of 0.1 Hz to 0.5 Hz during obstacle avoidance, and obtain a quantitative index reflecting the intensity of motion sickness by performing time-domain cumulative calculation on the vibration amplitude in this frequency range. When the quantitative indicator exceeds the preset comfort warning line, the intervention logic of the comprehensive evaluation feedback adjustment device is triggered. The stress parameter mapping processing device is also equipped with visual fatigue assessment logic. The visual fatigue assessment logic calculates the visual cognitive load of the virtual human body by extracting the refresh rate and contrast changes of the virtual display interface in the digital simulation space and the visual passing speed generated by the target movement, and uses the visual cognitive load as a supplementary component of the physiological stress parameters.

5. The simulation test system for the aircraft obstacle avoidance algorithm according to claim 4, characterized in that, The comprehensive evaluation feedback adjustment device establishes a set of multi-objective optimization evaluation criteria, which assign different weight coefficients to geometric obstacle avoidance safety, time optimality and physiological comfort. In the feedback regulation logic, the comprehensive evaluation feedback regulation device is configured to calculate the comprehensive cost value, which is equal to the weighted sum of the safety penalty item, the time cost item, and the physiological load item. Among them, the safety penalty item is inversely proportional to the minimum distance between the aircraft and the obstacle, the time cost item is directly proportional to the increased flight distance due to the avoidance action, and the physiological load item is composed of gravity overload, motion sickness index and visual fatigue. The comprehensive evaluation feedback adjustment device compares the comprehensive cost value of different potential avoidance paths in real time, generates optimized weights, and feeds them back to the obstacle avoidance algorithm logic processing device, so as to cause the obstacle avoidance trajectory to be dynamically corrected in a direction that conforms to physiological suitability while ensuring safe avoidance.

6. The simulation test system for the aircraft obstacle avoidance algorithm according to claim 5, characterized in that, The data monitoring and display device is deployed in a multi-screen display system or a virtual reality headset, and supports multi-channel synchronous playback. The multi-channel synchronous playback function is configured to display the first-view footage during obstacle avoidance with the dynamic curve of physiological stress parameters frame-aligned, so as to identify the specific obstacle avoidance stage that causes physiological discomfort to the personnel and the corresponding type of maneuver. The data monitoring and display device also includes an automatic report generation module. After the simulation test, the automatic report generation module is configured to automatically analyze the physiological data records during the entire obstacle avoidance process, mark all timestamps that violate the physiological comfort criteria, and extract the corresponding maneuver features for cluster analysis to identify defects in the logical design of the obstacle avoidance algorithm.

7. A simulation test method for aircraft obstacle avoidance algorithms, characterized in that, The simulation test system of the aircraft obstacle avoidance algorithm according to any one of claims 1-6 is used to carry out the simulation test of the aircraft obstacle avoidance algorithm.

Images