PathBench: A Benchmarking Platform for Classical and Learned Path Planning Algorithms

2021/05/04 by Alexandru-Iosif Toma, Hao-Ya Hsueh, Toma, Alexandru-Iosif +9 · 4 citations
Computer Science · Engineering · #Algorithm #Artificial intelligence #Autonomous Vehicle Technology and Safety #Benchmark (surveying) #Benchmarking #Computer science #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine learning #Motion planning #Path (computing) #Path length #Robot #Robotic Path Planning Algorithms #Robotics #Robotics (cs.RO) #Robotics and Sensor-Based Localization #cs.LG #cs.RO

paper · pdf · doi:10.48550/arxiv.2105.01777

published in arXiv (Cornell University) (Cornell University) · The Conference on Robots and Vision (CRV2021), Supplementary Website: https://sites.google.com/view/pathbench/

arxiv created 2021/05/04 · openalex publication_date 2021/05/04 · arxiv updated 2021/05/06 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/08

Abstract

Path planning is a key component in mobile robotics. A wide range of path planning algorithms exist, but few attempts have been made to benchmark the algorithms holistically or unify their interface. Moreover, with the recent advances in deep neural networks, there is an urgent need to facilitate the development and benchmarking of such learning-based planning algorithms. This paper presents PathBench, a platform for developing, visualizing, training, testing, and benchmarking of existing and future, classical and learned 2D and 3D path planning algorithms, while offering support for Robot Oper-ating System (ROS). Many existing path planning algorithms are supported; e.g. A*, wavefront, rapidly-exploring random tree, value iteration networks, gated path planning networks; and integrating new algorithms is easy and clearly specified. We demonstrate the benchmarking capability of PathBench by comparing implemented classical and learned algorithms for metrics, such as path length, success rate, computational time and path deviation. These evaluations are done on built-in PathBench maps and external path planning environments from video games and real world databases. PathBench is open source.

Citations

• Matplotlib: A 2D Graphics Environment
• Probabilistic roadmaps for path planning in high-dimensional configuration spaces
• Universal Planning Networks

Cited by

• Planetary Terrain Datasets and Benchmarks for Rover Path Planning
• Hybrid DQN-TD3 Reinforcement Learning for Autonomous Navigation in Dynamic Environments
• LLM-First Search: Self-Guided Exploration of the Solution Space
• Systematic Comparison of Path Planning Algorithms using PathBench

Related

• Systematic Comparison of Path Planning Algorithms using PathBench
• Neural Networks for Path Planning
• Motion Planning Networks: Bridging the Gap Between Learning-based and Classical Motion Planners
• Learning Human-Aware Path Planning with Fully Convolutional Networks
• Learning-based Fast Path Planning in Complex Environments