My research interest lies in algorithmic robotics and robot task and motion planning. Specifically I am actively deploying failure-explanation planning approaches, which reason about the failure of finding a valid plan and use the explanation for further guidance. Such approaches turn out to be very critical and effective to tackle challenging robotic manipulation tasks in complex and uncertain environment.
One project I am working on is to design algorithms which produce efficient and high-quality solutions for object rearrangement in cluttered and confined spaces such as shelves and fridges. The problem is much harder than the tabletop setup, as overhand grasps are not available and a robot must carefully maneuver to avoid undesirable robot-object interactions to fulfill a task. To tame the combinatorial challenges of the problem, I propose a more efficient monotone local solver, a high-quality non-monotone global planner and an effective pre-processing tool. For details of the project, refer to this github website.
Another project I had worked on, which featured failure-explanation planning, was to find safe and effective picking paths for robot arms in a setup where the target object is occluded by other surrounding objects or has only partial observability. I developed a probabilistic minimum constraint removal approach (MCR) to solve manipulation tasks with imperfect perception, where the uncertainty regarding objects are modelled as discrete distributions of object pose hypotheses. My planning solutions are capable of minimizing the probability of collision while maximizing the probability of reaching the target object. For more details of this research project, you can see this project website and its corresponding dataset.
Besides being committed to performing high-quality research, I also have great enthusiasm in outreach activities related to AI and robotics. I had been taking on the duty of course instructor for the advanced summer course Introduction to Artificial Intelligence for two consecutive years (2019, 2020). I had also contributed a book chapter covering how robotics can strengthen resiliency to disasters such as Covid-19 pandemic.
Prior to my Ph.D. in Rutgers, I earned my Master Degree in Mechanical Engineering from Columbia University in the city of New York, where I was working with Dr. Peter Allen on several robot car navigation tasks. I earned my Bachelor Degree in Vehicle Engineering from Nanjing University of Aeronautics and Astronautics, Nanjing, China.
For more information, feel free to contact me at rui [dot] wang26 [at] rutgers [dot] edu