Optimization of SIMD architectures for robot vision

Maximizing performance with a given constant cost or minimizing cost with a given constant performance by using either constant cost or constant performance as the optimization criterion is discussed. Simplified, well-approximated models of the cost and performance functions are considered and with these models a method is outlined to assess the benefit of variations in the configuration of the processors in a single-instruction multiple-data (SIMD) array based on the tradeoff between the number and the complexity of the processors. This method is applied to some aspects of the design, such as the datapath width and the reconfigurable capabilities