Obstacle detection method and device, unmanned aerial vehicle and storage medium

Abstract

The invention discloses an obstacle detection method and device, an unmanned aerial vehicle and a storage medium. The method comprises the following steps: acquiring pose data and a disparity map of an unmanned aerial vehicle, wherein the pose data and the disparity map have a corresponding relationship; determining pixel connectivity in a first specified direction in the disparity map according to the pose data, and extracting strip-shaped disparity features from the disparity map according to the pixel connectivity; clustering the strip parallax features to obtain clustering information; anddetermining obstacle information according to the clustering information. According to the technical scheme, the pose data of the unmanned aerial vehicle is utilized when the obstacle is detected, obstacle information drifting caused by movement of the unmanned aerial vehicle can be avoided, the whole process is low in complexity, consumption of computing resources is low, the obstacle detectiontime is shortened while the obstacle detection precision is guaranteed, and the invention is particularly suitable for equipment such as the unmanned aerial vehicle which conducts obstacle detection depending on the lightweight computing resources.

Description

technical field [0001] The present application relates to the technical field of obstacle avoidance, in particular to an obstacle detection method, device, unmanned aerial vehicle and storage medium. Background technique [0002] In the drone delivery business, the ability of the drone to avoid obstacles in the autonomous flight state is essential. In the UAV obstacle avoidance scheme, the binocular sensor has gradually become one of the main sensors in most UAV obstacle avoidance due to its advantages of low price, low power consumption, low weight, and rich obstacle information. At present, the methods of obstacle detection based on binocular sensors mostly directly use the disparity data or depth data calculated by binocular sensor matching, and directly convert them into three-dimensional space points as obstacles. However, the shortcomings of this type of method are also obvious, such as short detectable distance, too many false detections, etc., which are not conduciv...

AI Technical Summary

Problems solved by technology

However, the shortcomings of this type of method are also obvious, such as short detectable distance, too many false detections, etc., which are not conducive to practical application.

Images