A sensor fusion system using mapping learning method

Author

Mukai, Toshiharu ; Mori, Takashi ; Ishikawa, Masatoshi

Author_Institution

Fac. of Eng., Tokyo Univ., Japan

Volume

1

fYear

1993

fDate

26-30 Jul 1993

Firstpage

391

Abstract

A new method for sensor fusion is proposed and a sensor fusion system using the method with parallel processing architecture is shown. The method utilizes mappings between sensor data and can be applied not only to static coordination transforms between sensor data, but also to the dynamics of actuators. Capabilities of the method are confirmed by edge tracking experiments using the experimental sensor fusion system with a PSD (position sensitive detector), a CCD camera and an XY-recorder. The system is based on parallel processing architecture using transputers in a network. It is concluded that parallel processing is an intrinsic architecture for sensor fusion

Keywords

sensor fusion; CCD camera; XY-recorder; edge tracking experiments; mapping learning method; parallel processing architecture; position sensitive detector; sensor fusion system; transputers; Actuators; Charge coupled devices; Charge-coupled image sensors; Feedback loop; Image edge detection; Learning systems; Parallel processing; Robot sensing systems; Sensor fusion; Sensor systems;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Robots and Systems '93, IROS '93. Proceedings of the 1993 IEEE/RSJ International Conference on

Conference_Location

Yokohama

Print_ISBN

0-7803-0823-9

Type

conf

DOI

10.1109/IROS.1993.583127

Filename

583127