Towards Real-Time Vision-Based Manipulation of Workpieces

|The performance of real-time machine vision for workpiece manipulation is postulated to be describable by a sigmoid function of the computational resources made available for this purpose. Several attempts are discussed to estimate a representative resource reference value which characterizes the argument range where the hypothesized sigmoid function exhibits its steepest slope. Provided computational resources corresponding to at least the lower bound of the aforementioned range are made available, systematic experiments and even real applications become feasible. Additional computational resources for a combination of machine vision and robotics tend to become absorbed by other performance aspects than mere execution speed: a complex 5-dimensional optimization problem { comprising speed, robustness, versatility, analyzability, and maintainability { is outlined .