Abstract An experimental control system for the flexible transport of goods by convoys is described. In this transport system convoys can be built automatically. The goal is that every vehicle attach to or disengage from the convoy when it receives the corresponding order by remote control. Each trailer follows the previous one in the convoy by means of a specific vision module. The vision module consists of a low-cost hardware designed to acquire and binarize the image. The algorithms use this binary image to detect and locate a mark placed on the rear of the preceding trailer. From this information the guidance system defines the trajectory to be tracked and generates the control orders to the steering wheels. The trailers are automated guided vehicles (AGV), each having its own tracking system that does not require any physical contact.
[1]
James L. Crowley,et al.
Position estimation for a mobile robot using vision and odometry
,
1992,
Proceedings 1992 IEEE International Conference on Robotics and Automation.
[2]
Ernest L. Hall,et al.
Computer Image Processing and Recognition
,
1980,
IEEE Transactions on Pattern Analysis and Machine Intelligence.
[3]
W. von Seelen,et al.
Vision-based car-following: detection, tracking, and identification
,
1992,
Proceedings of the Intelligent Vehicles `92 Symposium.
[4]
Volker Graefe,et al.
Automatic recognition of vehicles approaching from behind
,
1992,
Proceedings of the Intelligent Vehicles `92 Symposium.
[5]
Michal Irani,et al.
Detecting and Tracking Multiple Moving Objects Using Temporal Integration
,
1992,
ECCV.