CN-122009517-A - Method for evaluating safety obstacle avoidance capability of unmanned aerial vehicle

CN122009517ACN 122009517 ACN122009517 ACN 122009517ACN-122009517-A

Abstract

The invention provides a method for evaluating the safety obstacle avoidance capability of an unmanned aerial vehicle, which comprises the steps of constructing an unmanned aerial vehicle obstacle avoidance track simulation test system, recording a real unmanned aerial vehicle movement track and a virtual expected track by the obstacle avoidance track simulation test system, and evaluating the safety obstacle avoidance capability of the unmanned aerial vehicle through comparative analysis. The system can reproduce air force and inertial force feeling in a real obstacle avoidance scene indoors in a controlled and repeatable environment, synchronously provide a test method of deceptive positioning signals, safely and accurately simulate the whole obstacle avoidance process of the unmanned aerial vehicle when the unmanned aerial vehicle encounters an obstacle outdoors, quantitatively evaluate the track quality of the unmanned aerial vehicle, and further provide data support for optimization and performance authentication of an obstacle avoidance algorithm.

Inventors

RONG ZHEN
HUANG XINHUI
CHEN LINJIAN
PAN XINHAO
CHENG WENMING
WANG PENG

Assignees

浣江实验室
浙江镭诺智能科技有限公司
深圳市雷诺智能技术有限公司
浙江航天润博测控技术有限公司
浙江弘飞空天科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260128

Claims (7)

1. A method of evaluating the safety obstacle avoidance capability of an unmanned aerial vehicle, comprising: The method comprises the steps of constructing an obstacle avoidance track simulation test system, wherein the obstacle avoidance track simulation test system comprises a physical test unit for building hardware equipment, a signal deception unit of a programmable GNSS signal simulator and a central control and simulation unit for a computer control system, the obstacle avoidance track simulation test system is cooperatively arranged through the physical test unit, the signal deception unit and the central control and simulation unit, and obstacle avoidance decisions of an unmanned aerial vehicle are reproduced indoors through integrated wind field simulation, GNSS signal deception and motion tracking technologies; After the test is finished, the obstacle avoidance track simulation test system records the real unmanned aerial vehicle movement track and the virtual expected track, and evaluates the unmanned aerial vehicle safety obstacle avoidance capacity through comparative analysis.
2. The method of claim 1, wherein the physical test unit comprises a test space, a multi-fan array wind field generator disposed within the test space, and a high-precision optical motion capture camera system; The multi-fan array wind field generator is used for generating a dynamically-changing simulated wind field in the test space according to the instruction; The high-precision optical motion capture camera system is used for tracking and feeding back the accurate position and the accurate gesture of the test unmanned aerial vehicle in the test space in real time.
3. The method of claim 2, wherein the signal spoofing unit generates control commands using a programmable GNSS signal simulator and transmits simulated satellite navigation signals to the test drone to misinterpret the drone as moving along a particular trajectory in an outdoor space that is in wide proximity to the drone.
4. The method for evaluating the safety obstacle avoidance capability of an unmanned aerial vehicle of claim 1, wherein the central control and simulation unit comprises: The obstacle scene simulation module is used for defining virtual obstacles and positions, sizes and motion tracks of the virtual obstacles in a virtual space; The unmanned aerial vehicle decision algorithm module is used for accessing and simulating an obstacle avoidance decision algorithm of the unmanned aerial vehicle to be tested; and the physical mapping and synchronous control module maps the virtual expected track generated by the unmanned aerial vehicle decision algorithm module into control instructions of the physical test unit and the signal spoofing unit in real time.
5. The method of claim 4, wherein the unmanned aerial vehicle decision algorithm module senses the virtual obstacle based on virtual position and speed information provided by the GNSS signal simulator and optionally other simulated sensor information and generates an expected obstacle avoidance trajectory command in real time.
6. The method for evaluating the safety obstacle avoidance capability of an unmanned aerial vehicle of claim 5, wherein the physical mapping and synchronization control module comprises the following working steps: a. the wind field mapping comprises the steps of calculating the relative wind speed born by the unmanned aerial vehicle in the current virtual state according to the real-time speed vector of the virtual expected track and the preset or virtual environment wind information, controlling the wind field generator to generate the real air flow corresponding to the relative wind speed, and acting on the indoor stationary or slowly drifting entity unmanned aerial vehicle to enable the body of the entity unmanned aerial vehicle to feel the same aerodynamic load as the outdoor flight; b. controlling the GNSS signal simulator to generate corresponding deception signals according to the real-time position and the speed of the virtual expected track, so that a flight control system of the unmanned aerial vehicle 'believes' to fly along the track; c. And dynamically synchronizing, namely receiving the real unmanned aerial vehicle position fed back by the motion capture camera system in real time, comparing the real unmanned aerial vehicle position with a virtual expected track, and dynamically fine-adjusting a wind field and GNSS signals through a closed-loop control algorithm to ensure that the real position of the real unmanned aerial vehicle is always constrained in a limited test space under the condition of overcoming real wind resistance, and simultaneously ensuring that the sensor experience of the real unmanned aerial vehicle is consistent with a virtual scene.
7. The method for evaluating the safety obstacle avoidance capability of an unmanned aerial vehicle of claim 3, wherein the high-precision optical motion capture camera system comprises a high-precision OptiTrack motion capture system and the GNSS signal simulator comprises a programmable vector signal generator.

Description

Method for evaluating safety obstacle avoidance capability of unmanned aerial vehicle Technical Field The invention relates to the technical field of unmanned aerial vehicle testing, in particular to a method for evaluating safety obstacle avoidance capability of an unmanned aerial vehicle. Background The prior art has realized that unmanned aerial vehicle is in the hover of indoor simulation uniform velocity flight and wind disturbance. However, the technology mainly aims at preset and linear flight tasks, and cannot meet the requirement of performing simulation test on the obstacle avoidance track of dynamic, non-preset and real-time decision. The obstacle avoidance process involves the unmanned aerial vehicle planning and executing complex tracks in real time according to sensor inputs, which puts higher demands on the dynamic response synchronism and the simulation authenticity of multiple physical fields of the test system. Therefore, development of a test method capable of reproducing air force and inertial force feeling in a real obstacle avoidance scene indoors and synchronously providing deceptive positioning signals is needed to quantitatively evaluate and optimize the obstacle avoidance track performance of the unmanned aerial vehicle. Disclosure of Invention The technical problem to be solved by the invention is to provide a method and a system for carrying out high-precision and repeatable simulation and evaluation on the obstacle avoidance track of an unmanned aerial vehicle in an indoor controlled environment by integrating wind field simulation, GNSS signal spoofing and motion tracking technologies, so as to solve the problems in the background technology. The technical problem solved by the invention is realized by adopting the following technical scheme that the method for evaluating the safety obstacle avoidance capability of the unmanned aerial vehicle comprises the following steps: The method comprises the steps of constructing an obstacle avoidance track simulation test system, wherein the obstacle avoidance track simulation test system comprises a physical test unit for building hardware equipment, a signal deception unit of a programmable GNSS signal simulator and a central control and simulation unit for a computer control system, the obstacle avoidance track simulation test system is cooperatively arranged through the physical test unit, the signal deception unit and the central control and simulation unit, and obstacle avoidance decisions of an unmanned aerial vehicle are reproduced indoors through integrated wind field simulation, GNSS signal deception and motion tracking technologies; After the test is finished, the obstacle avoidance track simulation test system records the real unmanned aerial vehicle movement track and the virtual expected track, and evaluates the unmanned aerial vehicle safety obstacle avoidance capacity through comparative analysis. As a further scheme of the invention: The physical test unit comprises a test space, a multi-fan array wind field generator arranged in the test space and a high-precision optical motion capture camera system; The multi-fan array wind field generator is used for generating a dynamically-changing simulated wind field in the test space according to the instruction; The high-precision optical motion capture camera system is used for tracking and feeding back the accurate position and the accurate gesture of the test unmanned aerial vehicle in the test space in real time. As a further scheme of the invention: The signal deception unit utilizes a programmable GNSS signal simulator to produce control instructions and transmits simulated satellite navigation signals to the test unmanned aerial vehicle, so that the unmanned aerial vehicle can mistakenly consider that the unmanned aerial vehicle is moving along a specific track in a wide outdoor space. As a further scheme of the invention: the central control and simulation unit comprises: The obstacle scene simulation module is used for defining virtual obstacles (static or dynamic) and positions, sizes and motion tracks of the virtual obstacles in a virtual space; The unmanned aerial vehicle decision algorithm module is used for accessing and simulating an obstacle avoidance decision algorithm of the unmanned aerial vehicle to be tested; and the physical mapping and synchronous control module maps the virtual expected track generated by the unmanned aerial vehicle decision algorithm module into control instructions of the physical test unit and the signal spoofing unit in real time. As a further scheme of the invention: The unmanned aerial vehicle decision algorithm module senses the virtual obstacle based on the virtual position and speed information provided by the GNSS signal simulator and optional other simulation sensor information and generates an expected obstacle avoidance track instruction in real time. As a further scheme of the invention: the physical mapping and synchronous