Developed and implemented a novel SOTA algorithm to attack the Placement Generation and Rearrangement Planning problems.
Created several ROS modules for the Baxter robot, including object recognition, visual servoing, pick-and-place, and real-time Bullet physics simulation.
Implemented SLAM on Lego MindStorms EV3.
Prepared and executed demos to encourage high-school students towards robotics and AI.
Produced presentations of SOTA cognitive robotics.
Conducted research on the object placement and rearrangement problems in the Cognitive Robotics group.
Implemented novel algorithms in the context of AI, mechatronics, and robotics.
GPA: 3.65 Thesis: Placement generation and hybrid planning for robotic rearrangement
on cluttered surfaces. Main Topics: Local Search, Hybrid Planning, Robotics.
In this thesis, we propose a general solution to the rearrangement of multiple
arbitrarily-shaped objects on a cluttered flat surface with multiple movable objects and obstacles. In
particular, we introduce a novel method to solve the object placement problem, utilizing nested local
searches guided by intelligent heuristics to efficiently perform multi-objective optimizations. The
solutions computed by our method satisfy the collision-freeness constraint, and involves minimal movements
of the clutter. Based on such a solution, we introduce a hybrid method to generate an optimal feasible
rearrangement plan, by integrating ASP-based high-level task planning with low-level feasibility checks. Our
hybrid planner is capable of solving challenging non-monotone rearrangement planning instances that cannot
be solved by the existing geometric rearrangement approaches.
A novel multifunctional laparoscopy tool intended for robotic-assisted surgery
is motorized and a control architecture using a set of simple PID controllers is developed.
Several tools to operate theBaxterrobot using its cameras to identify, localize
and provide visual servoing for pick-and-place operations. Also allows communication with the
Bullet physics engine via ROS to generate the environment it sees in live simulation.