Abstract
The performance demands in modern signal processing and control applications have motivated a trend towards utilisation of complex algorithms. This, in turn, has resulted in a resurgence in the development of high performance systems to make the real-time realisation of such algorithms feasible in practice. This paper presents an investigation into the performance of sequential architectures/uni-processor based systems in the real-time implementation of finite element algorithm of a flexible manipulator. The algorithm is implemented on a number of high-performance processors. A comparative assessment of the performance of the processors in implementing the algorithm, revealing the capabilities of the processors in relation to the nature of the algorithm, is presented in terms of execution time and speedup. The effect of compiler optimisation facilities on the execution time in implementing the algorithm is also demonstrated.
