CLeaR Lab, NUS
Date:
CLeAR lab in National University of Singapore seeks to improve people’s lives through intelligent robotics. The central focus of the lab is on developing physical and social skills for robots.
Role:
Implemented real life demonstration of LTLDoG, a diffusion based framework for satisfying LTL contraints. Using LTLDoG as a global planner we conducted experiments with a Unitree Go2 robot to demonstrate its effectiveness for real world navigation tasks. Our work was accepted in R-AL 2024 with the title LTLDoG: Satisfying Temporally-Extended Symbolic Constraints for Safe Diffusion-based Planning
Developed a simulation tool for autonmously generating sceaniors for social navigation tasks in Gazebo using ROS2. The framework utilizes the reasoning power of LLMs to generate elaborate scenarios with varying human behaviors to test and benchmark social navigations algorithims in differnt context and find failing edge cases. The work was presented at Unsolved Problems in Social Navigation workshop at RSS 2024. The short paper is titled Towards Automated Scenario Testing of Social Navigation Algorithms. Currently the work is submitted to ICRA 2025 and is under review.