A real-time vision guided vehicle-following system /
| Main Author: | |
|---|---|
| Other Authors: | , |
| Format: | Thesis Book |
| Language: | English |
| Published: |
1991.
|
| Subjects: | |
| Online Access: | ProQuest, Abstract Link to OAKTrust copy |
| Abstract: | For an autonomous vehicle to follow a lead vehicle, it needs to have the capability of measuring the position of the lead vehicle. In this study, we investigate a visually guided vehicle-following system. The system input is provided by a sequence of successive monocular images taken from a video camera mounted on the following vehicle. To avoid time-consuming image processing operations, a simple visual feature consisting of a light source on the lead vehicle is chosen to serve as a tracking feature. Based on the positions of the tracking point(lead vehicle), autonomous vehicle-following is achieved by obtaining a desired path such that a specified gap is maintained between the lead and following vehicles. Successive relative positions are used to define the lead vehicle's trajectory as a sequence of points. In addition, an automatic steering and a speed control scheme, requiring no inter-communication between vehicles, are presented and their performances are evaluated in terms of lateral and longitudinal errors. Finally, the proposed schemes are implemented on an autonomous vehicle developed at Texas A&M University, named BART. A number of experiments on realistic roadways has been carried out and the results are presented. The vehicle successfully drives on straight and curved roadways at speeds up to 20 mph even though there are substantial perturbations in measurements. |
|---|---|
| Item Description: | Typescript (photocopy). Vita. "Major subject: Electrical Engineering." |
| Physical Description: | xi, 96 leaves : illustrations ; 29 cm |
| Bibliography: | Includes bibliographical references. |