A vision system capable of extracting features from a semi-structured environment for vehicle guidance is described. It is primarily used for road following via the detection of mud tracks in the tropics. The scene captured by a CCD colour camera is digitised into 24-bit colour images with a resolution of 320x240 pixels. Partitioning of the scene into road and nonroad is based on the results of a colour image segmentation algorithm. The RGB images are converted to HSI format. Training samples are used to classify blocks of pixels using hue information. A Bayesian classifier in conjunction with a smooth thresholding function is used for segmentation. This approach results in traversable area recognition. Experimental results show that the algorithm is invariant to shadow conditions, i.e. roads were detected under varying fight conditions. Due to the soil conditions of the test sites, small puddles of water on the mud tracks are also classified as driveable areas. The system outputs a one bit 2D map of the image every 200 ms. Field results of the proposed approach have shown favourable responses for real-time implementation on an autonomous ground vehicle.
Real-time identification of driveable areas in a semi-structured terrain for an autonomous ground vehicle
2001
11 Seiten, 14 Quellen
Aufsatz (Konferenz)
Englisch