Title :
Fault-tolerant computing for robot kinematics using linear arithmetic codes
Author_Institution :
Dept. of Electr. & Comput. Eng., Illinois Inst. of Technol., Chicago, IL, USA
Abstract :
A fault-tolerant parallel processing method based on the linear arithmetic code is proposed for computation of robot kinematics. This method can detect and correct online a single error in the resultant matrix or vector with low hardware and time redundancy. The method also can be extended to multiple errors in computation. It can be applied to various hardware architectures for robot kinematics computation
Keywords :
fault tolerant computing; matrix algebra; parallel programming; robot programming; robots; fault-tolerant parallel processing method; hardware architectures; linear arithmetic codes; robot kinematics; time redundancy; Arithmetic; Computer architecture; Concurrent computing; Error correction; Fault tolerance; Hardware; Parallel processing; Redundancy; Robot kinematics; Vectors;
Conference_Titel :
Robotics and Automation, 1990. Proceedings., 1990 IEEE International Conference on
Conference_Location :
Cincinnati, OH
Print_ISBN :
0-8186-9061-5
DOI :
10.1109/ROBOT.1990.125988
