Unmanned driving foreign matter detection system and method

Abstract

The invention discloses an unmanned driving foreign matter detection system and method. The system comprises a track detection unit for shooting a track image and calibrating an effective detection area by tracking a track path in real time on the track image, an obstacle detection unit for shooting an image in front of a train in operation and identifying an obstacle through comparison with a standard track environment image, a radar detection unit for positioning a distance between the obstacle identified by the obstacle detection unit and the train, and a main control unit for carrying outgrading alarm on a train control center based on a detection result and providing a visual image of the obstacle. According to the detection system and method, multiple sensor technologies are fused for detection; a visual sensing device is used for replacing a driver to continuously see towards the front of the running train; obstacle features are extracted through an image identification technology, and the obstacle is positioned through a radar; and the obstacle can be detected and identified remotely, so that the response time is provided for early warning vehicle control and fault checking.

Description

technical field [0001] The invention relates to the technical field of rail transportation, in particular to an unmanned foreign object detection system and method. Background technique [0002] There are many underground tunnels in the subway environment, and the space is narrow and closed. The cables, bridges, pipelines, and electrical boxes on both sides of the tunnel may fall off and cause intrusion. Among them, abnormal phenomena such as obstacles on the rail surface, rockfall in the tunnel, and people falling pose a threat to driving safety. Serious, if it is not detected and dealt with in time, it is likely to cause serious accidents. [0003] At present, an obstacle detection system based on the principle of mechanical trigger is installed under the head of the train to replace the driver's function of looking at the line ahead. The travel rod moves, and then triggers the relay switch circuit, so that the train is stopped in an emergency. However, this method has o...

AI Technical Summary

Problems solved by technology

However, this method has obvious shortcomings, that is, it cannot realize early detection and early warning of obstacles in the track in the direction of the train, and improve driving safety. On the other hand, the detection of obstacles through mechanical collisions will inevitably cause damage to vehicle equipment.

[0004] In the prior art, remote obstacles within a certain angle range in front of the train are also detected by machine vision or laser radar, but this method cannot distinguish whether the obstacle belongs to the passageway of the vehicle

like figure 1 As shown, the detection system detects that there is a vehicle in front. In order to avoid rear-end collision, the detection system sends an alarm to the train control center to control the vehicle braking. However, if the vehicle under test is not on the path of the vehicle, it will form a false alarm and affect the normal operation.

Images