Improving Local Trajectory Optimisation using Probabilistic Movement Primitives

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 as the demonstrations. We demonstrate the effectiveness of our approach in several complex obstacle avoidance scenarios.

Authors

• R. B. Ashith Shyam
• Peter Lightbody
• Gautham P. Das
• Pengcheng Liu 0005
• Sebastián Gómez-González
• Gerhard Neumann

Keywords

• Costs
• Shape
• Probabilistic logic
• End effectors
• Trajectory
• Planning
• Collision avoidance
• Optimization
• Tuning
• Intelligent robots
• Local Optimum
• Trajectory Optimization
• Movement Primitives
• Probabilistic Movement Primitives
• Cost Function
• Local Minima
• Path Planning
• Human Experts
• Obstacle Avoidance
• Distribution Of Trajectories
• Inverse Reinforcement Learning
• Collision-free Trajectory
• Total Cost
• Parameter Space
• Urban Planning
• Weight Vector
• Block Diagonal
• Mean Vector
• Goal State
• Joint Space
• Rapidly-exploring Random Tree
• Starting State
• Task Space
• Smooth Trajectory
• Robot Operating System
• Trajectory Planning
• Robot Body
• Trajectories In Space
• Geometric Primitives
• motion planning
• gradient optimisation
• robot manipulation
• learning from demonstrations.