Dexterous Manipulation Primitives for the Real Robot Challenge
This work addresses robotic manipulation tasks for real-world applications, but it is incremental as it builds on existing primitives and optimization methods.
The authors tackled the problem of solving cuboid manipulation tasks in the Real Robot Challenge by decomposing tasks into primitives like grasping, turning, and re-positioning, using model-based trajectory optimization and a state-machine for sequencing, resulting in their team placing second in the challenge.
This report describes our approach for Phase 3 of the Real Robot Challenge. To solve cuboid manipulation tasks of varying difficulty, we decompose each task into the following primitives: moving the fingers to the cuboid to grasp it, turning it on the table to minimize orientation error, and re-positioning it to the goal position. We use model-based trajectory optimization and control to plan and execute these primitives. These grasping, turning, and re-positioning primitives are sequenced with a state-machine that determines which primitive to execute given the current object state and goal. Our method shows robust performance over multiple runs with randomized initial and goal positions. With this approach, our team placed second in the challenge, under the anonymous name "sombertortoise" on the leaderboard. Example runs of our method solving each of the four levels can be seen in this video (https://www.youtube.com/watch?v=I65Kwu9PGmg&list=PLt9QxrtaftrHGXcp4Oh8-s_OnQnBnLtei&index=1).