Active obstacle detection system based on unmanned driving

Abstract

The invention provides an active obstacle detection system based on unmanned driving. The active obstacle detection system comprises an obstacle detection module located in a cab and two-door anti-pinch platforms located in the middles of the two sides of each compartment. The obstacle detection module comprises a digital high-definition camera and an infrared high-definition camera, and the two cameras synchronously shoot images in real time and send the images to the video processing server for foreign matter invasion detection. The video processing server uses two cameras to shoot an imagepair, obtains three-dimensional information of the target object through solving two-dimensional information of the target object in the image pair, and carries out foreign matter determination and foreign matter invasion risk degree discrimination. The two-door anti-pinch platform comprises hemispherical high-definition cameras arranged on the two sides of a train door, the hemispherical high-definition cameras take pictures of the area between the train door and a shielding door after stopping and send the pictures to the video processing server, the video processing server carries out foreign matter recognition on the ROI area in the two-door space, and therefore the situation that the train is still started under the condition that foreign matter is pinched is avoided.

Description

technical field [0001] The invention relates to unmanned active obstacle detection and belongs to the technical field of rail transit vehicle control. Background technique [0002] Unmanned subway trains have special requirements for the detection of obstacles in front of the train and the anti-pinch function of the second door, and need to give alarms and prompts in the driver's cab and OCC according to the application scenarios, and store and upload videos / images to OCC at the same time. OCC remotely retrieves on-site images. [0003] After searching, it was found that the Chinese invention patent application CN 109188460 A discloses an unmanned foreign object detection system and method. The system includes: a track detection unit, which captures track images, and calibrates the effective detection area by tracking track tracks in real time on the track images The obstacle detection unit is used to capture the image in front of the train, and identify obstacles by compar...

AI Technical Summary

Problems solved by technology

[0005] However, it uses a single visible light camera to capture images in front of the train, and its effective range is relatively short, and it must be combined with radar to determine whether there is an obstacle. Often, an alarm can only be issued after the obstacle approaches, leaving a short time for the system to respond

Images