|
Accession number;02A0585367
|
| Title;Recognizing Moving Objects for Robot Navigation using Realtime Omnidirectional Sereo Vision. |
| Author;
KOYASU HIROSHI
(Osaka Univ.)
MIURA JUN
(Osaka Univ.)
SHIRAI YOSHIAKI
(Osaka Univ.)
|
Journal Title;Nippon Kikai Gakkai Robotikusu, Mekatoronikusu Koenkai Koen Ronbunshu
|
Journal Code:L0318A
|
ISSN:
|
|
VOL.2001;NO.Pt.3;PAGE.2A1.B5(1)-2A1.B5(2)(2001)
|
| Figure&Table&Reference;FIG.5, REF.4 |
| Pub. Country;Japan |
| Language;Japanese |
| Abstract;This paper proposes a method to recognize moving objects in a wide view and in real-time, such a recognition is needed for mobile robots to move in a dynamic environment. Our method uses a stereo vision of two omnidirectional cameras to obtain omnidirectional range data in real-time, and first makes a grid map of the surrouding environment of the robot. Compareing range data with the map, robot can detect moving objects and can estimates their position and velocity with Kalman filter by integrating sequence of the detected position. We demonstrate the effectiveness of our method by an experiment. (author abst.) |
|
|
|
Related Articles;
|
