Enhancing Student Learning in Robot Path Planning Optimization through Graph-Based Methods

Abstract

Optimizing robot path planning within computational intelligence and robotics is increasingly important, and graph-based models are at the forefront of this advancement.Teaching these complex subjects poses challenges, addressed in this study through a novel pedagogical approach that combines sparrow-dissection and scaffolding with active, project-based learning (SDS-AL).This method, implemented in a graduate Computational Intelligence course, centers on teaching a visibility graph-based model for robot path planning.Students are provided with source code, which they dissect, understand, and adapt for their specific projects.The course structure encourages students to progress from guided learning to independent project completion, enhancing their practical and theoretical understanding of graph-based techniques.The efficacy of this teaching method is assessed through milestone assignments, presentations, and student feedback, focusing on their comprehension and application of the concepts.Feedback is also gathered on their development and use of neural network models, based on the revision of the initial source code.The teaching strategies are aligned with the course's learning outcomes, confirmed by analyzing the students' projects on graph-based robot path planning.The effectiveness of the course is further evaluated by integrating this analysis with data from the course evaluation system, underscoring the successful application of the graph-based method and the high quality of learning achieved.This integrated approach, merging sparrow-dissection and scaffolding pedagogies with active, project-based involvement, significantly improves student comprehension and application skills in the domain of robot path planning optimization.