Automatic programming of a time-optimal robot controller and an analog electrical circuit to implement the robot controller by means of genetic programming

Abstract

Genetic programming is an automatic programming technique that evolves computer programs to solve, or approximately solve, problems. This paper presents two examples in which genetic programming creates a computer program for controlling a robot so that the robot moves to a specified destination point in minimal time. In the first approach, genetic programming evolves a computer program composed of ordinary arithmetic operations and conditional operations to implement a time-optimal control strategy. In the second approach, genetic programming evolves the design of an analog electrical circuit consisting of transistors, diodes, resistors, and power supplies to implement a near-optimal control strategy.