High-dimensional Motion Planning using Latent Variable Models via Approximate Inference

In this work, we present an efficient framework to generate a motion trajectory of a robot that has a high degree of freedom (e.g., a humanoid robot). High-dimensionality of the robot configuration space often leads to difficulties in utilizing the widely-used motion planning algorithms because the volume of the decision space increases exponentially with the number of dimensions. To handle complications arising from the large decision space, and to solve a corresponding motion planning problem efficiently, two key concepts were adopted. First, the Gaussian process latent variable model (GPLVM) was utilized for low-dimensional representation of the original configuration space. Second, an approximate inference algorithm was used, exploiting through the duality between control and estimation, to explore the decision space and to compute a high-quality motion trajectory of the robot. Utilizing the GP-LVM and the duality between control and estimation, we construct a fully probabilistic generative model with which we transform a high-dimensional motion planning problem into a tractable inference problem. Finally, we compute the optimal motion trajectory via an approximate inference algorithm based on a variant of the particle filter. Numerical experiments are presented to demonstrate the applicability and validity of the proposed approach. The resulting motions can be viewed in the supplemental video.