Improving Local Trajectory Optimisation using Probabilistic Movement Primitives
Shyam, RB Ashith
MetadataShow full item record
Local trajectory optimisation techniques are a powerful tool for motion planning. However, they often get stuck in local optima depending on the quality of the initial solution and consequently, often do not find a valid (i.e.collision free) trajectory. Moreover, they often require fine tuning of a cost function to obtain the desired motions. In this paper, we address both problems by combining local trajectory optimisation with learning from demonstrations. The human expert demonstrates how to reach different target end-effector locations in different ways. From these demonstrations, we estimate a trajectory distribution, represented by a Probabilistic Movement Primitive (ProMP). For a new target location, we sample different trajectories from the ProMP and use these trajectories as initial solutions for the local optimisation. As the ProMP generates versatile initial solutions for the optimisation,the chance of finding poor local minima is significantly reduced.Moreover, the learned trajectory distribution is used to specify the smoothness costs for the optimisation, resulting in solutions of similar shape than the demonstrations. We demonstrate the effectiveness of our approach in several complex obstacle avoidance scenarios.
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) November 4-8, 2019, Macau, China;
Shyam, R.B.A., Lightbody, P., Das, G., Liu, P., Gomez-Gonzalez, S. and Neumann, G. (2019) 'Improving Local Trajectory Optimisationusing Probabilistic Movement Primitives', IEEE/RSJ International Conference on Intelligent Robots and Systems, November 4 - 8, Macau, China.
Paper presented at IEEE/RSJ International Conference on Intelligent Robots and Systems, November 4 - 8, 2019, Macau, China, available at: https://doi.org/10.1109/IROS40897.2019.8967980.
Cardiff Metropolitan University (Grant ID: Cardiff Metropolian (Internal))
Showing items related by title, author, subject and abstract.
Geometric analysis-based trajectory planning and control for underactuated capsule systems with viscoelastic property Liu, Pengcheng; Yu, Hongnian; Cang, Shuang (Sage, 2017-09-20)This paper proposes a novel geometric analysis-based trajectory planning approach for underactuated capsule systems with viscoelastic property. The idea is to reduce complexity and to characterize coupling by imposing a ...
Trajectory Synthesis and Optimization of an Underactuated Microrobotic System with Dynamic Constraints and Couplings Liu, Pengcheng; Yu, Hongnian; Cang, Shuang (Springer, 2018-07-25)Motivated by the desire to optimally control the friction-induced stick-slip locomotion and sufficiently improve the energy efficacy, a novel trajectory synthesis and optimization scheme is proposed in this paper for a ...
Liu, Pengcheng; Yu, Hongnian; Cang, Shuang (Springer, 2018-07-07)This paper studies the issue of adaptive trajectory tracking control for an underactuated vibro-driven capsule system and presents a novel motion generation framework. In this framework, feasible motion trajectory is derived ...