Exploring Optimal Parameter Combinations for Genetic Programming in Robot Trajectory Generation

This work explores the relevance of appropriately selecting execution parameters in generating efficient routes for mobile robots in static environments using genetic programming (GP). The experiments conducted in this study focused on determining the optimal execution parameters and analyzing the algorithm’s behavior in different scenarios. The results highlight the importance of adapting the execution parameters to the specific problem being addressed. The experiments show that larger populations and more generations are not always necessary for achieving satisfactory results. GP faces challenges finding optimal solutions in scenarios with dispersed food sources due to the increased search space dimensionality and multiple clusters. Moreover, a trade-off between computational effort and solution quality is observed, emphasizing the need to evaluate the parameter settings critically. This research highlights the importance of parameter adaptation in GP. It provides insights into reducing computational effort without compromising solution quality, thus enhancing the practical applicability of GP in real-world robotics and intelligent transportation systems.