UAV obstacle avoidance device and obstacle avoidance method based on millimeter wave radar and vision fusion

Abstract

A drone obstacle avoidance device based on millimeter wave radar and vision fusion, installed on the drone, where the radar module is used to emit millimeter waves and receive millimeter waves reflected by obstacles within the detection range of the drone After obtaining the data sampled at N points, send it to the information fusion processing module; the vision module is used to obtain the obstacle visual image data within the detection range of two different drones at time t, and then send it to the information fusion processing module; the electronic compass The module is used to obtain the three-dimensional inclination information of the UAV, and send it to the information fusion processing module; the information fusion processing module is used to obtain the three-dimensional information of obstacles within the detection range of the UAV according to the data sampled at N points, and then send it to the flight Control module; the flight control module is used to avoid obstacles according to the three-dimensional information of obstacles within the detection range of the drone; t represents a time-varying variable.

Description

technical field [0001] The invention belongs to the field of UAV obstacle avoidance applications, in particular to an UAV obstacle avoidance device and obstacle avoidance method based on millimeter wave radar and vision fusion, which are suitable for obtaining three-dimensional information of obstacles. Background technique [0002] Radar actively emits electromagnetic waves and receives reflected waves of targets within the detection range to detect targets, and the radar has a long range, high reliability, and is not affected by weather, haze, etc., and has the advantage of working all-weather and all-day. With the rapid development of millimeter-wave chips, the size and weight of millimeter-wave radars are smaller, and the resolution and detection accuracy are getting higher and higher; however, a single millimeter-wave radar can only obtain the distance information of obstacles, but cannot obtain the orientation information of obstacles. . [0003] The current mainstrea...

AI Technical Summary

Problems solved by technology

[0002] Radar actively emits electromagnetic waves and receives reflected waves of targets within the detection range to detect targets, and the radar has a long range, high reliability, and is not affected by weather, haze, etc., and has the advantage of working all-weather and all-day. With the rapid development of millimeter-wave chips, the size and weight of millimeter-wave radars are smaller, and the resolution and detection accuracy are getting higher and higher; however, a single millimeter-wave radar can only obtain the distance information of obstacles, but cannot obtain the orientation information of obstacles.

[0003] The current mainstream method of UAV obstacle avoidance is to use optical sensors. A single optical sensor can only obtain the orientation information of obstacles, but cannot obtain the distance information of obstacles; while the system using multiple optical sensors can obtain three-dimensional information of obstacles. information, but in the data processing stage, the processor performance is very high. It is difficult for the current embedded processor to complete the data processing tasks of multiple optical sensors in real time.

[0004] In the current method based on the fusion of millimeter-wave radar and visual sensor, multiple millimeter-wave radars are usually used to obtain three-dimensional information of obstacles, and visual sensors play an auxiliary role, or multiple optical sensors are used to obtain three-dimensional information of obstacles The radar plays an auxiliary role, but the fusion of multiple optical sensors has heavy data processing tasks, which increases the requirements of the obstacle avoidance system for the processor performance, and cannot guarantee the flight safety of the UAV.

Images