Abstract
This brief proposes a novel stochastic method that exploits the particular kinematics of mechanisms with redundant actuation and a well-known manipulability measure to track the desired end-effector task-space motion in an efficient manner. Whilst closed-form optimal solutions to maximise manipulability along a desired trajectory have been proposed in the literature, the solvers become unfeasible in the presence of obstacles. A manageable alternative to functional motion planning is thus proposed that exploits the inherent characteristics of null-space configurations to construct a generic solution able to improve manipulability along a task-space trajectory in the presence of obstacles. The proposed Stochastic Constrained Optimization (SCO) solution remains close to optimal whilst exhibiting computational tractability, being an attractive proposition for implementation on real robots, as shown with results in challenging simulation scenarios, as well as with a real 7R Sawyer manipulator, during surface conditioning tasks.
Highlights
Mapping the path of an end-effector onto a configuration trajectory for the robot to accomplish a desired collision-free task is a well-known problem in robotics [1]
Increasing the manipulability of the robotic system at each time step is regarded as an effective means of moving away from the neigborhood of such configurations [3], reducing the hazardous condition whereby small task space movements may translate to large joint velocities
Handles kinematic restrictions on the end effector exactly and does not rely on a discretization of the task space or configuration space; Exploits the null-space at each configuration along the path to maximize the manipulability of the robot while avoiding obstacles; Produces smooth trajectories that can be directly commanded to the robot without the need for a posterior smoothing phase; Delivers a fitting experimental performance for challenging motion problems
Summary
Mapping the path of an end-effector onto a configuration trajectory for the robot to accomplish a desired collision-free task is a well-known problem in robotics [1]. The consideration of redundancy, where the actuated degrees of freedom of the manipulator exceed the end-effector variables defining its functionality in the task space, adds an interesting dimension to the planning problem It effectively facilitates a scheme where additional objectives can be incorporated along the way. Operating away from singularity regions relaxes the effect of undesirable dynamics that otherwise impose additional perturbances to the robot controllers, permitting superior end-effector precision whilst executing the desired task. This brief proposes a stochastic method that exploits the particular kinematics of closed-chain mechanisms with redundant actuation and a well-known manipulability measure [4] to track the desired end-effector task-space motion in an efficient manner.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have