Zherong Pan1,
Min Liu2,4, and
Xifeng Gao3, and
Dinesh Manocha4
Department of Computer Science, University of North Carolina at Chapel Hill1
Department of Computer Science, National University of Defense Technology2
Department of Computer Science, Florida State University3
Department of Computer Science, Nankai University4
We present a new algorithm for computing globally optimal topology and trajectory jointly for 2D planar linkages. Planar linkage structures can generate complex end-effector trajectories using only a single rotational actuator, which is very useful in building low-cost robots. We address the problem of searching for the optimal topology and geometry of these structures and present new optimization methods that consider topology changes that are non-smooth and non-differentiable. We formulate this problem as a mixed-integer convex programming (MICP) problem for which a global optimum can be found using the branch-and-bound (BB) algorithm. As a result, within a finite amount of time, our method can find planar linkage structures with end-effector trajectories that closely match the user-specified target trajectories. We tested our method to search for planar linkages with 5-7 rigid bodies. Compared with sampling-based methods or simulated annealing, our method improves the quality of the solution by at most 7X and the optimized planar linkage structure has been tested on a 4-legged walking robot.
Globally Optimal Joint Search of Topology and Trajectory for Planar Linkages
Zherong Pan, Min Liu, Xifeng Gao, and Dinesh Manocha
International Symposium on Robotics Research (ISRR 2019) [Paper]
Download Video: [MP4]