Obstacle avoidance method and device for agriculture and forestry robot, computer equipment and storage medium

Abstract

The invention relates to an obstacle avoidance method and device for an agriculture and forestry robot, electronic equipment and a storage medium. The method comprises the steps that an image acquisition device on the agriculture and forestry robot is started, and an image processing model is preloaded; when the agriculture and forestry robot performs agriculture and forestry operation in a targetscene, image data of the surrounding environment of the agriculture and forestry robot is acquired through the image acquisition device; the image data is processed through the image processing model, and target agriculture and forestry assets in the image data are recognized; the image processing model is obtained through training based on data in a terrain simulation model; based on a target area where the target agriculture and forestry assets are located in the image data, a first distance between the target agriculture and forestry assets in the surrounding environment of the agricultureand forestry robot and the agriculture and forestry robot is determined; and when the first distance is smaller than or equal to a preset distance, the agriculture and forestry robot is controlled tomove in the direction away from the target agriculture and forestry assets. By the adoption of the method, the obstacle avoidance capability of the agriculture and forestry robot can be improved.

Description

technical field [0001] The present application relates to the field of image recognition, in particular to an agricultural and forestry robot obstacle avoidance method, device, computer equipment and storage medium. Background technique [0002] With the development of agricultural and forestry mechanization and digitization, agricultural and forestry robots are gradually emerging. Agricultural and forestry robots mostly use remote control or autonomous travel mode to operate. In disaster relief, irrigation, fertilization and other scenarios, agricultural and forestry robots are more and more used. [0003] In the existing scheme, the agricultural and forestry robot cannot actively judge the properties of the surrounding objects during the operation. The agricultural and forestry robot travels according to the route set in advance by the program or manually, and detects obstacles through infrared sensors or laser beds to prevent collisions. . However, when using sensors to...

AI Technical Summary

Problems solved by technology

[0003] In the existing scheme, the agricultural and forestry robot cannot actively judge the properties of the surrounding objects during the operation. The agricultural and forestry robot travels according to the route set in advance by the program or manually, and detects obstacles through infrared sensors or laser beds to prevent collisions.

However, when using sensors to detect obstacles, inaccurate detection often occurs. For example, infrared sensors can easily misjudge black areas, and the laser emitted by infrared sensors is absorbed in black areas, so the device cannot judge whether the front can travel; The laser emitted by the sensor can penetrate glass, which is easy to misjudge translucent objects

Such inaccurate recognition may easily lead to the problem that the agricultural and forestry robot's obstacle avoidance of obstacles is not timely enough.

Images