An unmanned helicopter perception and obstacle avoidance simulation system

Abstract

The application provides an unmanned helicopter sensing and obstacle avoidance simulation system, which comprises a flight control computer, an avionics simulation machine, a model computer, an obstacle avoidance sensing and path planning computer, a load simulation computer, a rudder system, a helicopter control device, a visual simulation computer, a visual display system and a display control system. By using the application, a real flight test field environment can be simulated in a laboratory condition, and a semi-physical simulation verification of an obstacle avoidance sensing sensor and an obstacle avoidance algorithm can be completed.

Description

Technical Field

[0001] This invention relates to a perception and obstacle avoidance simulation system for unmanned helicopters. Background Technology

[0002] With the development of drone technology, its functions are becoming increasingly rich and its applications are becoming more and more widespread. To improve drone flight safety in complex environments, the need for research and practical application of real-time drone perception and obstacle avoidance is becoming increasingly urgent. Because drones themselves and their onboard equipment are expensive, relying on frequent flight tests for testing and verification is risky and time-consuming. Therefore, it is necessary to research corresponding obstacle avoidance simulation systems that reproduce three-dimensional images in virtual space and combine them with superimposed two-dimensional curve information to observe the drone's flight status in the three-dimensional scene at any time, minimizing the number of actual test flights and identifying potential equipment problems and software defects as early as possible.

[0003] Most current drone obstacle avoidance simulations are conducted through digital simulations, lacking a complete obstacle avoidance simulation system that can combine highly realistic 3D flight scenarios, flight dynamics models, obstacle avoidance sensors, and obstacle avoidance algorithms to enable drones to fly in simulated real-world environments. Summary of the Invention

[0004] Purpose of the invention: The technical problem to be solved by the present invention is to provide an unmanned helicopter perception and obstacle avoidance simulation system to address the shortcomings of the existing technology, and to verify the flight performance of the unmanned helicopter in a simulated real environment.

[0005] The system of this invention includes a flight control computer, an avionics simulator, a model computer, an obstacle avoidance perception and path planning computer, a load simulation computer, a rudder system, helicopter control equipment, a visual simulation computer, a visual display system, and a display and control system;

[0006] The flight control computer receives helicopter control commands via a serial port, receives flight status feedback information sent from the avionics simulator via a serial port, controls the flight of the unmanned helicopter, and sends servo control quantities to the avionics simulator via a serial port.

[0007] The avionics simulator has a real-time simulation function. It communicates with the model computer via a network, feeds back the collected unmanned helicopter status information to the flight control computer, and sends control signals to the model computer as input to the model. The avionics simulator communicates with the rudder system via a serial port and sends control quantities to the rudder.

[0008] The model computer runs a UAV simulation model in real time, communicates with the avionics simulator through a network, receives control signals sent by the avionics simulator, and feeds back the UAV helicopter status information to the avionics simulator after model calculation. At the same time, the model status information is sent to the visual display system for real-time display through the network.

[0009] The load simulation computer is used to simulate various physical characteristics of airborne mission load equipment in a three-dimensional virtual space, realize obstacle avoidance sensor simulation, and calculate visual collision data.

[0010] The obstacle avoidance perception and path planning computer receives real-time detection information from obstacle avoidance sensors via a network, calculates obstacle targets, performs path planning, and sends obstacle avoidance waypoints to the flight control computer via a serial port to enable the UAV to avoid obstacles during flight. The obstacle avoidance sensors are installed in the payload simulation computer, including lidar simulation, millimeter-wave radar simulation, etc.

[0011] The rudder system includes a rudder controller and a rudder. The rudder controller receives control signals sent by the avionics simulator via a serial port, drives the rudder to move, and feeds back the actual position information of the rudder to the avionics simulator.

[0012] The helicopter control equipment sends internal control commands and external control handle inputs to the flight control computer via a serial port.

[0013] The visual simulation computer constructs a three-dimensional task scene and receives model computer data through the network to drive the three-dimensional motion of the scene's unmanned helicopter.

[0014] The visual display system is used to project and display three-dimensional mission scenes and the movement status of drones;

[0015] The display and control system communicates with the avionics equipment simulator, model computer, obstacle avoidance perception and path planning computer, load simulation computer, and steering system via network to control, display, analyze, and store the status of each device in the simulation system.

[0016] The obstacle avoidance perception and path planning computer executes the obstacle avoidance algorithm, specifically including: receiving real-time detection information from obstacle avoidance sensors via the network, calculating obstacle targets, and planning paths. During simulated flight, the obstacle avoidance perception and path planning computer first processes point cloud data, including point cloud filtering and noise reduction, then performs point cloud segmentation. Based on point cloud intensity, spatial location, and geometric information, it divides the obstacle area of ​​the UAV's flight environment, establishes a point cloud map of obstacles and the surrounding environment, and calculates a safe path around the obstacle using obstacle information described by the grid map, combining the current relative position with the obstacle and the aircraft's motion state. Finally, it smooths the path, simplifies it to waypoints on the obstacle avoidance path, and sends them to the flight control computer for obstacle avoidance route tracking control. After flying through the entire obstacle avoidance waypoint until no obstacle is detected, it transitions to normal simulated flight.

[0017] The obstacle avoidance path tracking control specifically includes: the flight control computer performing obstacle avoidance path tracking control based on the helicopter's flight attitude, speed, altitude information, and obstacle avoidance waypoint information, enabling the helicopter to track the obstacle avoidance waypoints and achieve the obstacle avoidance process, expressed as:

[0018]

[0019] Among them, K x K is the proportional control parameter for the forward position deviation from the target point. y It is the proportional control parameter for the lateral position deviation from the target point, x c Here, x is the forward position deviation setpoint, y is the forward position deviation feedback value, and y is the forward position deviation setpoint. c y is the lateral position deviation setting value, y is the lateral position deviation feedback value, and V is the lateral position deviation setting value. x It is the forward speed setpoint, V y This is the lateral speed setting value;

[0020] The forward speed and lateral speed are calculated by proportional-integral-derivative control to obtain the longitudinal pitch control quantity and lateral pitch control quantity of the helicopter, respectively. When the helicopter flies close to the target obstacle avoidance point within X1 meters, it automatically switches to the next target obstacle avoidance point until the entire obstacle avoidance waypoint is completed.

[0021] The simulation process of the system includes the following steps:

[0022] Step 1: Power on the system;

[0023] Step 2: Configure the unmanned helicopter flight mission, including aircraft type settings, mission scenario settings, flight environment settings, obstacle settings, mission payload configuration, obstacle avoidance algorithm configuration, and run the simulation system;

[0024] Step 3: Send flight control commands for the unmanned helicopter through the helicopter control equipment, including automatic takeoff, hovering, and flight path.

[0025] Step 4: Monitor the simulation process data through the display and control system, including the flight status of the unmanned helicopter, the operating status of the onboard sensors, the operating status of the rudder system, and the obstacle perception status.

[0026] Step 5: Monitor the attitude, speed and position of the unmanned helicopter in the three-dimensional scene through the visual display system, including the interaction status between the unmanned helicopter and the mission scene. The mission scene includes terrain, airport, houses, mountains, forests and man-made obstacles.

[0027] Step 6: When performing perception and obstacle avoidance simulation tasks, if the unmanned helicopter successfully completes the flight mission, the simulation ends, and the stored data is analyzed and the effect is evaluated; if the UAV collides with the target and crashes, the stored data is analyzed, and steps 2 to 5 are repeated to modify the mission payload configuration and obstacle avoidance algorithm configuration, and the simulation is restarted.

[0028] In step 5, the unmanned helicopter performs the following flight effects during the perception and obstacle avoidance simulation tasks:

[0029] The inadequate performance of the sensing sensors prevented the unmanned helicopter from detecting the target, resulting in a collision and crash.

[0030] The perception sensor performance was sufficient, but the obstacle avoidance algorithm was unsuitable. The output obstacle avoidance path could not guide the unmanned helicopter to avoid the target in time, causing the unmanned helicopter to collide with the target and crash.

[0031] With sufficient sensor performance and a suitable obstacle avoidance algorithm, the output obstacle avoidance path can guide the unmanned helicopter to avoid the target in time, and the unmanned helicopter can successfully complete the flight mission.

[0032] Step 6, the analysis and effect evaluation of the stored data, includes:

[0033] By analyzing the drone's flight attitude, speed, and position, we determine whether it is flying normally. If it is flying normally, we further examine the obstacle avoidance performance, including whether the attitude, speed, and position are smooth and the minimum distance to obstacles, in order to correct the obstacle avoidance algorithm. If it crashes, we further analyze the cause of the crash to determine whether the obstacle avoidance sensor or the obstacle avoidance algorithm is inappropriate.

[0034] In step 6, the stored data is analyzed to evaluate the obstacle avoidance flight performance. The evaluation objects include the helicopter's attitude response amplitude, response quickness, attitude control deviation, and obstacle avoidance waypoint tracking control deviation. The pitch angle quickness expression is as follows:

[0035] q pk / Δθ pk (2)

[0036] The shortcut expression for roll angle is as follows:

[0037] p pk / Δφ pk (3)

[0038] Where, q pk It is the peak value of the pitch angular velocity response, Δθ pk It is the peak value of the pitch attitude angle response, p pk It is the peak value of the roll angular velocity response, Δφ pk It is the peak value of the roll attitude angle response;

[0039] Attitude control deviation is measured by standard deviation and must not exceed design requirements. The expression is:

[0040]

[0041]

[0042] Where, θ i It is the pitch angle calculated in real time, where i takes values ​​from 1 to n. It is the average pitch angle of n frames, S θ It is the standard deviation of pitch angle control. It is the roll angle calculated in real time. It is the average roll angle of n beats. It is the standard deviation of roll angle control;

[0043] The obstacle avoidance waypoint tracking control deviation is measured by the circular probability Δl error and must not exceed the design requirements. The expression is:

[0044]

[0045]

[0046] Δl=0.5887×(Δy_rms+Δh_rms) (8)

[0047] Where, Δy i It is the lateral offset calculated in real time, Δy_rms is the standard deviation of the lateral offset in n frames, and Δh i It is the height difference calculated in real time, and Δh_rms is the standard deviation of the height difference for n frames.

[0048] Beneficial effects: In the research process of UAV perception and obstacle avoidance technology, it is necessary to first conduct semi-physical simulation tests to ensure the effectiveness of the perception sensors and obstacle avoidance algorithms used before carrying out field flight tests.

[0049] Using this invention, the real flight test environment can be simulated under laboratory conditions to complete the hardware-in-the-loop simulation verification of obstacle avoidance perception sensors and obstacle avoidance algorithms. The effectiveness of obstacle avoidance sensor types and parameters can be verified separately, the effectiveness of obstacle avoidance algorithms can be verified separately, and the effectiveness of the combination of obstacle avoidance sensors and obstacle avoidance algorithms can be verified. During the project development process, especially in the prototype stage, it guides technical R&D personnel to conduct sufficient technical exploration and verification. Attached Figure Description

[0050] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, and the advantages of the present invention in the above and / or other aspects will become clearer.

[0051] Figure 1 This is a diagram of the simulation system of the present invention.

[0052] Figure 2 This is a schematic diagram of collision detection under the visual environment of the present invention.

[0053] Figure 3 This is a flowchart of the obstacle avoidance and path planning process of the present invention.

[0054] Figure 4 This is a schematic diagram of obstacle avoidance simulation data for the present invention.

[0055] Figure 5 This is the laser point cloud data map of the present invention.

[0056] Figure 6 This is a schematic diagram illustrating the obstacle avoidance effect of the present invention. Detailed Implementation

[0057] This invention provides an unmanned helicopter perception and obstacle avoidance simulation system, including a flight control computer, an avionics simulator, a model computer, an obstacle avoidance perception and path planning computer, a payload simulation computer, a rudder system, helicopter control equipment, a visual simulation computer, a visual display system, and a display and control system. The system composition of this invention is as follows: Figure 1 As shown.

[0058] The flight control computer runs flight control software in real time, receives helicopter control commands via serial port, receives flight status feedback information sent from the avionics simulator via serial port, controls the flight of the unmanned helicopter, and sends servo control quantities to the avionics simulator via serial port.

[0059] The avionics simulator runs real-time simulation software for the unmanned helicopter's onboard sensors, control systems, and telemetry and control links, providing real-time simulation capabilities for the interface characteristics, dynamic characteristics, and fault characteristics of each sensor and control unit. The simulator communicates with the model computer via a network, feeding back the collected unmanned helicopter status information to the flight control computer and sending control signals to the model computer as input. The simulator also communicates with the control system via a serial port, sending control signals to the control units.

[0060] The model computer runs a UAV simulation model in real time, communicates with the avionics simulator via a network, receives control signals sent by the avionics simulator, and feeds back the UAV helicopter status information to the avionics simulator after model calculation; at the same time, the model status information is sent to the visual display system for real-time display via the network.

[0061] The load simulation computer is used to simulate various physical characteristics of airborne mission load equipment in a three-dimensional virtual space, realize load simulation such as lidar simulation and millimeter-wave radar simulation, and calculate visual collision data.

[0062] The lidar simulation includes parameters such as radar detection range, horizontal field of view, vertical field of view, horizontal angular resolution, vertical angular resolution, frame rate, rotation speed, and point cloud data output rate. The type and main parameters of the lidar can be selected and configured through the simulation control console in the display and control system. The lidar point cloud data output is as follows: Figure 5 As shown.

[0063] The millimeter-wave radar simulation includes parameters such as radar detection range, range resolution, target velocity, velocity resolution, horizontal field of view, vertical field of view, and angular resolution. The type and main parameters of the millimeter-wave radar can be selected and configured through the simulation control console in the display and control system.

[0064] The collision detection process is as follows: Figure 2 As shown, a simplified model is used for calculation. Terrain, city buildings, airports, and target objects are represented using a whiteboard. A camera model is used, and the positions of the current eye point and the predicted eye point of the camera form a line segment. It is then determined whether this line segment intersects with the model.

[0065] The obstacle avoidance perception and path planning computer runs obstacle avoidance and path planning algorithms. It receives information such as point cloud data detected by radar through the network, performs obstacle detection and feature extraction, and calculates waypoints for obstacle avoidance planning based on flight dynamics constraints. The waypoints are then sent to the flight control computer via serial port to enable the UAV to avoid obstacles during flight.

[0066] The obstacle avoidance and path planning algorithm flow is as follows: Figure 3As shown, during normal simulated flight, the obstacle avoidance sensor detects obstacles injected into the simulated environment. The obstacle avoidance perception and path planning computer first processes the point cloud data, including point cloud filtering and noise reduction. Then, it performs point cloud segmentation, dividing the obstacle region of the UAV's flight environment based on point cloud intensity, spatial location, and geometric information, creating a point cloud map of the obstacles and surrounding environment. Combining the current relative position to the obstacle and the aircraft's motion state, it calculates a safe path to bypass the obstacle using obstacle information described by the grid map. Finally, the path is smoothed, simplifying the waypoints on the obstacle avoidance path and sending them to the flight control computer for obstacle avoidance route tracking control. After flying through all obstacle avoidance waypoints until no more obstacles are detected, the flight transitions to normal simulated flight.

[0067] The obstacle avoidance path tracking control involves the flight control computer performing obstacle avoidance path tracking control based on the helicopter's flight attitude, speed, altitude information, and obstacle avoidance waypoint information. This allows the helicopter to track the obstacle avoidance waypoints and achieve the obstacle avoidance process. Its expression is:

[0068]

[0069] Among them, K x K is the proportional control parameter for the forward position deviation from the target point. y It is the proportional control parameter for the lateral position deviation from the target point, x c x is the forward position deviation setting value, x is the forward position deviation feedback value, and y is the forward position deviation setting value. c y is the lateral position deviation setting value, y is the lateral position deviation feedback value, and V is the lateral position deviation setting value. x It is the forward speed setpoint, V y This is the lateral speed setpoint. The forward speed and lateral speed are calculated by PID control to obtain the longitudinal pitch control value and the lateral pitch control value of the helicopter, respectively. When the helicopter flies close to the target obstacle avoidance point X1 (5 in this embodiment) within a certain range, it automatically switches to the next target obstacle avoidance point until it has flown through the entire obstacle avoidance waypoint.

[0070] The rudder system consists of rudder controllers and rudders, among other airborne components. The rudder controller receives control signals from the avionics simulator via a serial port, drives the rudders to move, and feeds back the actual position information of the rudders to the avionics simulator.

[0071] The helicopter control equipment sends internal control commands and external control handle inputs to the flight control computer via a serial port.

[0072] The visual simulation computer constructs a three-dimensional task scene and receives model computer data via a network to drive the three-dimensional motion of the scene's unmanned helicopter platform.

[0073] The three-dimensional mission scenario includes a simulated terrain database and a three-dimensional model database. The terrain database is based on the actual simulation scenario, with key areas created using real photographs, and includes models of terrain, rivers, lakes, mountains, etc.; the three-dimensional model database is developed based on the actual shape and structure of the unmanned helicopter and mission payload.

[0074] The visual display system is used to project and display three-dimensional mission scenes and the movement status of drones.

[0075] The display and control system communicates with the avionics simulator, model computer, obstacle avoidance perception and path planning computer, payload simulation computer, and rudder system via network to control, display, analyze, and store the status of each device in the simulation system. The display and control system includes a simulation master control console, a mission payload monitoring console, a ground flight control monitoring console, a model monitoring console, an avionics simulation monitoring console, a rudder monitoring console, and an analysis monitoring console.

[0076] The data analysis is used to evaluate the obstacle avoidance flight performance, including helicopter attitude response amplitude, response speed, attitude control deviation, and obstacle avoidance waypoint tracking control deviation. During obstacle avoidance flight, the attitude response amplitude must not exceed the helicopter's safe flight envelope; and the response speed must be good, meaning the helicopter has the ability to quickly change attitude and rapidly avoid obstacles. Its expression is:

[0077] Pitch angle speedup: q pk / Δθ pk (2)

[0078] Roll angle speedup: p pk / Δφ pk (3)

[0079] Where, q pk It is the peak value of the pitch angular velocity response, Δθ pk It is the peak value of the pitch attitude angle response, p pk It is the peak value of the roll angular velocity response, Δφ pk It is the peak value of the roll attitude angle response.

[0080] Attitude control deviation is measured by standard deviation and should not exceed the design requirements. Its expression is:

[0081]

[0082]

[0083] Where, θ i The pitch angle is calculated in real time, i = 1 to n. It is the average pitch angle of n frames, S θ It is the standard deviation of pitch angle control. It is the roll angle calculated in real time, i = 1 to n. It is the average roll angle of n beats. It is the standard deviation of the roll angle control.

[0084] The obstacle avoidance waypoint tracking control deviation is measured by the circular probability Δl error and should not exceed the design requirements. Its expression is:

[0085]

[0086]

[0087] Δl=0.5887×(Δy_rms+Δh_rms) (8)

[0088] Where, Δy i The lateral offset is calculated in real time, i = 1 to n, Δy_rms is the standard deviation of the lateral offset in n beats, and Δh i It is the height difference calculated in real time, i = 1 to n, and Δh_rms is the standard deviation of the height difference in n frames.

[0089] The simulation control console runs simulation control software, including a visual scene status control module, a task scene configuration module, and an algorithm configuration module.

[0090] The visual scene status control module includes functions such as: time, function switching, wind direction, rain, snow, fog and cloud simulation control, observation position setting, and viewing angle setting.

[0091] The mission scenario configuration module includes functions such as mission scenario settings, UAV model settings, flight environment settings, obstacle settings, and mission payload configuration.

[0092] The algorithm configuration module includes functions such as obstacle avoidance algorithm configuration.

[0093] In this invention, the obstacle avoidance simulation diagram is as follows: Figure 4 The simulation process includes the following steps:

[0094] Step 1: Power on the system;

[0095] Step 2: Configure the unmanned helicopter flight mission through the simulation control software, including aircraft type settings, mission scenario settings, flight environment settings, obstacle settings, mission payload configuration, obstacle avoidance algorithm configuration, etc., and run the simulation system.

[0096] Step 3: Send flight control commands for the unmanned helicopter through the helicopter control equipment, including automatic takeoff, hovering, and flight path.

[0097] Step 4: Monitor the simulation process data through the display and control system, including the flight status of the unmanned helicopter, the operating status of the onboard sensors, the operating status of the rudder system, the obstacle perception status, and the obstacle avoidance route planning status.

[0098] Step 5: Monitor the operational status of the unmanned helicopter in the three-dimensional scene through the visual display system, especially its interaction with the mission scene, which includes terrain, airport, houses, mountains, forests and other man-made obstacles.

[0099] When performing perception and obstacle avoidance simulation tasks, there may be various flight effects, including but not limited to:

[0100] (1) The sensor performance was insufficient and the target could not be detected, resulting in the unmanned helicopter colliding with the target and crashing;

[0101] (2) The sensor performance is sufficient, but the obstacle avoidance algorithm is not suitable. The output obstacle avoidance path cannot guide the unmanned helicopter to avoid the target in time, resulting in the unmanned helicopter colliding with the target and crashing.

[0102] (3) The sensor performance is sufficient and the obstacle avoidance algorithm is appropriate. The output obstacle avoidance path can guide the unmanned helicopter to avoid the target in time, and the unmanned helicopter can successfully complete the flight mission.

[0103] Step 6: During the perception and obstacle avoidance simulation task, if the unmanned helicopter successfully completes the flight mission, the simulation ends, and the stored data is analyzed and the effect is evaluated. If the UAV collides with the target and crashes, the stored data is analyzed, and steps 2 to 5 are repeated, modifying the mission payload configuration and obstacle avoidance algorithm configuration, and the simulation is restarted.

[0104] The obstacle avoidance effect diagram of this invention is shown below. Figure 5 , Figure 6 As shown in the figure, the obstacle avoidance simulation process can be seen from the figure. When the lidar detects the obstacle on the right in real time, it will output the lidar point cloud data. The unmanned helicopter will automatically deviate from the original preset route, maintain a safe distance from the obstacle, and carry out obstacle avoidance flight.

[0105] This invention provides a simulation system for perception and obstacle avoidance of unmanned helicopters. Many methods and approaches exist for implementing this technical solution; the above description is merely a preferred embodiment of the invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications should also be considered within the scope of protection of this invention. All components not explicitly stated in this embodiment can be implemented using existing technologies.

Claims

1. A simulation system for perception and obstacle avoidance of an unmanned helicopter, characterized in that, This includes flight control computer, avionics simulator, model computer, obstacle avoidance perception and path planning computer, load simulation computer, rudder system, unmanned helicopter control equipment, visual simulation computer, visual display system, and display and control system; The flight control computer receives unmanned helicopter control commands via serial port, receives flight status feedback information sent by avionics simulator via serial port, controls the flight of unmanned helicopter, and sends servo control quantities to avionics simulator via serial port. The avionics simulator has a real-time simulation function. It communicates with the model computer via a network, feeds back the collected unmanned helicopter status information to the flight control computer, and sends control signals to the model computer as input to the model. The avionics simulator communicates with the rudder system via a serial port and sends control quantities to the rudder. The unmanned helicopter simulation model runs in real time in the model computer, communicates with the avionics equipment simulator through the network, receives control signals sent by the avionics equipment simulator, and feeds back the unmanned helicopter status information to the avionics equipment simulator after model calculation. At the same time, the model status information is sent to the visual display system for real-time display through the network. The load simulation computer is used to simulate various physical characteristics of airborne mission load equipment in a three-dimensional virtual space, realize obstacle avoidance sensor simulation, and calculate visual collision data. The obstacle avoidance perception and path planning computer receives real-time detection information from the obstacle avoidance sensor via the network, calculates the obstacle target, performs path planning, and sends the obstacle avoidance waypoints to the flight control computer via serial port to achieve obstacle avoidance flight of the unmanned helicopter; the obstacle avoidance sensor is installed in the load simulation computer. The rudder system includes a rudder controller and a rudder. The rudder controller receives control signals sent by the avionics simulator via a serial port, drives the rudder to move, and feeds back the actual position information of the rudder to the avionics simulator. The unmanned helicopter control equipment sends internal control commands and external control handle inputs to the flight control computer via a serial port. The visual simulation computer constructs a three-dimensional task scene and receives model computer data through the network to drive the three-dimensional motion of the scene's unmanned helicopter. The visual display system is used to project and display three-dimensional mission scenes and the motion status of unmanned helicopters. The display and control system communicates with the avionics equipment simulator, model computer, obstacle avoidance perception and path planning computer, load simulation computer, and steering system via network to control, display, analyze, and store the status of each device in the simulation system.

2. The system according to claim 1, characterized in that, The obstacle avoidance perception and path planning computer executes the obstacle avoidance algorithm, specifically including: receiving real-time detection information from obstacle avoidance sensors via the network, calculating obstacle targets, and planning paths. During simulated flight, the obstacle avoidance perception and path planning computer first processes point cloud data, including point cloud filtering and noise reduction, then performs point cloud segmentation. Based on point cloud intensity, spatial location, and geometric information, it divides the obstacle area of ​​the unmanned helicopter's flight environment, establishes a point cloud map of obstacles and the surrounding environment, and calculates a safe path around the obstacle using obstacle information described by the grid map, combining the current relative position with the obstacle and the aircraft's motion state. Finally, it smooths the path, simplifies it to waypoints on the obstacle avoidance path, and sends them to the flight control computer for obstacle avoidance route tracking control. After flying through the entire obstacle avoidance waypoint until no obstacle is detected, it transitions to normal simulated flight.

3. The system according to claim 2, characterized in that, The obstacle avoidance path tracking control specifically includes: the flight control computer performing obstacle avoidance path tracking control based on the unmanned helicopter's flight attitude, speed, altitude information, and obstacle avoidance waypoint information, enabling the unmanned helicopter to track obstacle avoidance waypoints and achieve the obstacle avoidance process, expressed as: Among them, K x K is the proportional control parameter for the forward position deviation from the target point. y It is the proportional control parameter for the lateral position deviation from the target point, x c Here, x is the forward position deviation setpoint, y is the forward position deviation feedback value, and y is the forward position deviation setpoint. c y is the lateral position deviation setting value, y is the lateral position deviation feedback value, and V is the lateral position deviation setting value. x It is the forward speed setpoint, V y This is the lateral speed setting value; The forward speed and lateral speed are calculated by proportional-integral-derivative control to obtain the longitudinal pitch control quantity and lateral pitch control quantity of the unmanned helicopter, respectively. When the unmanned helicopter flies close to the target obstacle avoidance point within X1 meters, it automatically switches to the next target obstacle avoidance point until the entire obstacle avoidance waypoint is completed.

4. The system according to claim 3, characterized in that, The simulation process of the system includes the following steps: Step 1: Power on the system; Step 2: Configure the unmanned helicopter flight mission, including aircraft type settings, mission scenario settings, flight environment settings, obstacle settings, mission payload configuration, obstacle avoidance algorithm configuration, and run the simulation system; Step 3: Send flight control commands for the unmanned helicopter, including automatic takeoff, hovering, and flight path, through the unmanned helicopter control equipment; Step 4: Monitor the simulation process data through the display and control system, including the flight status of the unmanned helicopter, the operating status of the onboard sensors, the operating status of the rudder system, and the obstacle perception status. Step 5: Monitor the attitude, speed and position of the unmanned helicopter in the three-dimensional scene through the visual display system, including the interaction status between the unmanned helicopter and the mission scene. The mission scene includes terrain, airport, houses, mountains, forests and man-made obstacles. Step 6: When performing the perception and obstacle avoidance simulation task, if the unmanned helicopter successfully completes the flight mission, the simulation ends, and the stored data is analyzed and the effect is evaluated; if the unmanned helicopter collides with the target and crashes, the stored data is analyzed, and steps 2 to 5 are repeated to modify the mission payload configuration and obstacle avoidance algorithm configuration, and the simulation is restarted.

5. The system according to claim 4, characterized in that, In step 5, the unmanned helicopter performs the following flight effects during the perception and obstacle avoidance simulation tasks: The inadequate performance of the sensing sensors prevented the unmanned helicopter from detecting the target, resulting in a collision and crash. The perception sensor performance was sufficient, but the obstacle avoidance algorithm was unsuitable. The output obstacle avoidance path could not guide the unmanned helicopter to avoid the target in time, causing the unmanned helicopter to collide with the target and crash. With sufficient sensor performance and a suitable obstacle avoidance algorithm, the output obstacle avoidance path can guide the unmanned helicopter to avoid the target in time, and the unmanned helicopter can successfully complete the flight mission.

6. The system according to claim 5, characterized in that, Step 6, the analysis and effect evaluation of the stored data, includes: By monitoring the flight attitude, speed, and position of the unmanned helicopter, it is determined whether the flight is normal. If the flight is normal, the obstacle avoidance effect is further examined, including whether the attitude, speed, and position are smooth and the minimum distance to the obstacle, in order to correct the obstacle avoidance algorithm. If the helicopter crashes, the cause of the crash is further analyzed to determine whether the obstacle avoidance sensor or the obstacle avoidance algorithm is inappropriate.

7. The system according to claim 6, characterized in that, In step 6, the stored data is analyzed to evaluate the obstacle avoidance flight performance. The evaluation objects include the unmanned helicopter's attitude response amplitude, response speed, attitude control deviation, and obstacle avoidance waypoint tracking control deviation. The pitch angle speed expression is as follows: q pk / Dth pk (2) The shortcut expression for roll angle is as follows: p pk / Df pk (3) Where, q pk It is the peak value of the pitch angular velocity response, Δθ pk It is the peak value of the pitch attitude angle response, p pk It is the peak value of the roll angular velocity response, Δφ pk It is the peak value of the roll attitude angle response; Attitude control deviation is measured by standard deviation and must not exceed design requirements. The expression is: Where, θ i It is the pitch angle calculated in real time, where i takes values ​​from 1 to n. It is the average pitch angle of n frames, S θ It is the standard deviation of pitch angle control. It is the roll angle calculated in real time. It is the average roll angle of n beats. It is the standard deviation of roll angle control; The obstacle avoidance waypoint tracking control deviation is measured by the circular probability Δl error and must not exceed the design requirements. The expression is: Δl=0.5887×(Δy_rms+Δh_rms) (8) Where, Δy i It is the lateral offset calculated in real time, Δy_rms is the standard deviation of the lateral offset in n frames, and Δh i It is the height difference calculated in real time, and Δh_rms is the standard deviation of the height difference for n frames.

Images