Machine Learning-Driven Robotic Path Planning: A Synthesis of Classical and Modern Approaches

DOI

10.2991/978-94-6463-823-3_45How to use a DOI?

Keywords

Machine Learning-Driven Robotics; Path Planning Optimization; Deep Reinforcement Learning

Abstract

This review synthesizes advancements in ML-driven robotic path planning, integrating classical and modern approaches. Millán and Torras’ reinforcement connectionist framework established early foundations for continuous action space learning, later expanded by deep reinforcement learning (DRL) architectures. Recent innovations focus on hybrid optimization frameworks, multi-robot collaboration, and generalization capabilities. Quantitative evaluations demonstrate 15–40% shorter paths, 25–95% faster convergence, and 89–95% obstacle avoidance success. Notable implementations include Long et al., where CNN-DRL fusion led to a 30% reduction in path length and 90% success in obstacle avoidance, facilitated by obstacle-aware navigation. Nippun Kumaar and Kochuvila enhanced mobile robots via reward function optimization and experience replay, attaining 40% path efficiency and 89% obstacle success. Wang et al. developed a globally guided multi-robot system with 25% faster convergence and 95% avoidance rates, maintaining scalability across dynamic environments. Emerging trends highlight algorithm hybridization (e.g., combining DRL with metaheuristics for improved efficiency), edge computing deployment for real-time processing, and optimization of energy-time trade-offs in constrained environments. These developments position ML-driven planners as critical for next-gen autonomous systems, particularly in dynamic scenarios requiring real-time adaptability. The field’s evolution highlights progression from theoretical frameworks to practical implementations addressing computational efficiency, multi-agent coordination, and environmental uncertainty mitigation.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Kai Feng
PY  - 2025
DA  - 2025/08/31
TI  - Machine Learning-Driven Robotic Path Planning: A Synthesis of Classical and Modern Approaches
BT  - Proceedings of the 2025 3rd International Conference on Image, Algorithms, and Artificial Intelligence (ICIAAI 2025)
PB  - Atlantis Press
SP  - 458
EP  - 467
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-823-3_45
DO  - 10.2991/978-94-6463-823-3_45
ID  - Feng2025
ER  -