Collaborative, Learning, and Adaptive Robots Lab at NUS.
We Develop Physical and Social Intelligence for Trustworthy Robots.
We introduce PhysiCLeaR, an annotated dataset of everyday objects and tactile readings collected from a Gelsight Mini sensor, as well as Octopi, a system that leverages both tactile representation learning and large vision-language models to perform physical reasoning and inference, given tactile videos of multiple objects.
We develop interpolant policies that leverage informative source distributions for imitation learning.
We introduce the third iteration of the Arena platform - Arena 3.0, a platform to develop, train, and benchmark social navigation approaches!
We propose an optimization-based grasp synthesis framework, GRaCE, to generate context-specific grasps in complex scenarios. We test GRaCE in a simulator and a real-world grasping tasks.
Harold gave a well-received Keynote at ICRA’24 on our group’s work on generative modeling and robotics! He spoke about two of our recent works on new kinds o...
All four of our submissions (3 papers, 1 demo) were accepted to R:SS 2024! A fantastic accomplishment by our CLeAR members Kaiqi, Samson, Tasbolat, Linh, Kel...
Both our submitted papers were accepted to Neurips this year! Come join us in New Orleans! Congrats to Alvin and Shuyue. Find out more about the papers below...
CLeAR had three papers accepted to IROS this year! Come join us in Detroit! Congrats to Tasbolat, Bowen, and Kaiqi. Find out more about the papers below: ...
We show that a single unconditional diffusion model performs competitively in out-of-distribution detection tasks by measuring the rate-of-change and curvature of diffusion paths connecting data samples to the standard normal distribution.
We develop a safe planning method for trajectory generation by sampling from diffusion model under different LTLf constraints.
We extend gradient flow methods to a variety of high-quality image synthesis tasks using a novel density ratio learning method.
We introduce a novel framework centered around the Probable Object Location (POLo) score, which allows the agent to make data-driven decisions for efficient object search.
