Obstacle detection method, inspection robot vehicle-mounted obstacle avoidance system and method

Abstract

The invention relates to an obstacle detection method and an inspection robot vehicle-mounted obstacle avoidance system and method. An inspection robot is provided with a detection radar system and an image acquisition system. The detection radar system comprises a plurality of detection radars for synchronously detecting obstacle information, and the detection radars are synchronously started to detect the obstacle information; and when any detection radar detects obstacle information or the inspection robot runs to a curve road section, the image acquisition system is synchronously started to acquire obstacle pictures according to the obstacle information detected by the detection radar system, and effective target obstacle information is approved and recognized. According to the invention, the detection radar system and the image acquisition system are combined to realize judgment and detection of obstacles together, so that the detection precision of the obstacles is improved, and the intelligent obstacle avoidance requirement of the subway tunnel inspection robot is met.

Description

technical field [0001] The invention belongs to the technical field of rail transit inspection, and in particular relates to an obstacle detection method, a vehicle-mounted obstacle avoidance system and method of an inspection robot. Background technique [0002] The inspection robot in the existing subway tunnel uses a single detection module to detect obstacles, but now there are many interference objects on both sides of the tunnel in some tunnel sections, such as electrical cabinets, pipelines, warning signs, traffic lights, etc., the single detection module Recognition has a high rate of false positives in this environment. At the same time, the existing detection modules mainly use infrared ranging, ultrasonic ranging or visual system identification, but the intelligent obstacle avoidance system using infrared ranging and ultrasonic ranging cannot visually observe the appearance and size of obstacles after the detection is completed. etc.; the intelligent obstacle avo...

AI Technical Summary

Problems solved by technology

[0002] The inspection robot in the existing subway tunnel uses a single detection module to detect obstacles, but now there are many interference objects on both sides of the tunnel in some tunnel sections, such as electrical cabinets, pipelines, warning signs, traffic lights, etc., the single detection module Recognition has a high rate of false positives in this environment

At the same time, the existing detection modules mainly use infrared ranging, ultrasonic ranging or visual system identification, but the intelligent obstacle avoidance system using infrared ranging and ultrasonic ranging cannot visually observe the appearance and size of obstacles after the detection is completed. etc.; the intelligent obstacle avoidance system using visual system recognition, its visual recognition system is relatively large, and it is not easy to be carried on the tunnel inspection robot

[0003] For example, the patent CN206773526U wheeled robot composite intelligent obstacle avoidance system uses a combination of ultrasonic ranging module and infrared ranging module to detect obstacles, but the ultrasonic ranging module has a very short measurement distance, and distant obstacles cannot be measured. The infrared ranging module measures distant obstacles alone, the recognition rate is low, and it cannot accurately warn, and the intelligent obstacle avoidance system cannot adapt well to the subway tunnel environment

Patent CN108268036A is a new robot intelligent obstacle avoidance system. This system uses multiple ultrasonic sensors and infrared sensors to detect the distance, shape and speed of obstacles respectively. More impact, high false positive rate

Images