Optimization of a Proportional-Summation-Difference Controller for a Line-Tracing Robot Using Bacterial Memetic Algorithm

The smart home of the future will require a universal and dynamic mapping system of a complex labyrinth that devices can use for a wide range of tasks, including item delivery, monitoring and streamlining transportation. In this paper, the core concept of this system is proposed and is demonstrated by a line-tracing robot that has self optimizing properties using an evolutionary algorithm. Tests were performed to find the best controller for the robot and a proportional-summation-difference (PSD) controller was found to be best suited for the robot’s hardware specifications. An evolutionary computing algorithm called Bacterial Memetic Algorithm (BMA) was then implemented to optimize the PSD controller’s properties to suite the conditions of a track. After several thousand candidate solutions of evolutionary computing and simulated tests, the robot was able to complete a lap in real time with a significant decrease in both lap time and average error.