HyperPlan: A Framework for Motion Planning Algorithm Selection and Parameter Optimization

Over the years, many motion planning algorithms have been proposed. It is often unclear which algorithm might be best suited for a particular class of problems. The problem is compounded by the fact that algorithm performance can be highly dependent on parameter settings. This paper shows that hyperparameter optimization is an effective tool in both algorithm selection and parameter tuning over a given set of motion planning problems. We present different loss functions for optimization that capture different notions of optimality. The approach is evaluated on a broad range of scenes using two different manipulators, a Fetch and a Baxter. We show that optimized planning algorithm performance significantly improves upon baseline performance and generalizes broadly in the sense that performance improvements carry over to problems that are very different from the ones considered during optimization.

Authors

• Mark Moll
• Constantinos Chamzas
• Zachary Kingston
• Lydia E. Kavraki

Keywords

• Manipulators
• Planning
• Optimization
• Tuning
• Portfolios
• Intelligent robots
• Guidelines
• Selection Algorithm
• Algorithm Parameters
• Path Planning
• Planning Algorithm
• Motion Planning Algorithms
• Loss Function
• Tuning Parameter
• Class Of Problems
• Hyperparameter Tuning
• Baseline Performance
• Planning Problem
• Focus Of This Paper
• Loss Value
• Problem Instances
• Planning Time
• Bayesian Optimization
• Starting State
• Robot Motion
• Time Budget
• Cost Path
• Asymptotic Optimality
• Algorithm Configuration
• Rounds Of Optimization
• Start Of Section
• Mobile Manipulator
• Portfolio Selection
• Hyperparameter Space
• Worst-case Performance
• Solution Path
• Loss Function Design