What is it about?
SocNavBench: is a simulated benchmark that will run your social navigation algorithm through a gauntlet of scenarios and compare vs popular baselines! SocNavBench comes with scenarios containing pre-recorded humans being replayed, and handpicked (or randomly assigned) robot starts & goals. It also has a suite of metrics that measure the safety and efficiency of your navigation. Our code is public: https://github.com/CMU-TBD/SocNavBench
Featured Image
Photo by Lukas on Unsplash
Read the Original
This page is a summary of: SocNavBench: A Grounded Simulation Testing Framework for Evaluating Social Navigation, ACM Transactions on Human-Robot Interaction, September 2022, ACM (Association for Computing Machinery),
DOI: 10.1145/3476413.
You can read the full text:
Resources
Contributors
The following have contributed to this page
