Motion planning through policy search
نویسندگان
چکیده
We propose a motion planning algorithm for performing policy search in the full pose and velocity space of a mobile robot. By comparison, existing techniques optimize high-level plans, but fail to optimize the low-level motion controls. We use policy search in a high dimensional control space to find plans that lead to measurably better motion planning. Our experimental results suggest that our approach leads to superior robot motion than many existing techniques.
منابع مشابه
Importance sampling-based approximate optimal planning and control
In this paper, we propose a sampling-based planning and optimal control method of nonlinear systems under non-differentiable constraints. Motivated by developing scalable planning algorithms, we consider the optimal motion plan to be a feedback controller that can be approximated by a weighted sum of given bases. Given this approximate optimal control formulation, our main contribution is to in...
متن کاملGenerating State Lattice Motion Primitives for Differentially Constrained Motion Planning
We propose a method of generating motion primitives for motion planning under differential constraints. These primitives lend themselves particularly well to off-line precomputation. In this manner, they are attractive due to their capacity to capture the mobility constraints of the robot and to establish a state sampling policy that is conducive to efficient search. The first objective allows ...
متن کاملLearning an Interface to Improve Efficiency in Combined Task and Motion Planning
In mobile manipulation planning, it is not uncommon for tasks to require thousands of individual motions. Planning complexity is exponential in the length of the plan, rendering direct motion planning intractable for many problems of interest. Recent work has focused on task and motion planning (TAMP) as a way to address this challenge. TAMP methods integrate logical search with continuous geom...
متن کاملData-driven Planning via Imitation Learning
Robot planning is the process of selecting a sequence of actions that optimize for a task specific objective. For instance, the objective for a navigation task would be to find collision free paths, while the objective for an exploration task would be to map unknown areas. The optimal solutions to such tasks are heavily influenced by the implicit structure in the environment, i.e. the configura...
متن کاملPolicy Search for Imitation Learning
Efficient motion planning and possibilities for non-experts to teach new motion primitives are key components for a new generation of robotic systems. In order to be applicable beyond the well-defined context of laboratories and the fixed settings of industrial factories, those machines have to be easily programmable, adapt to dynamic environments and learn and acquire new skills autonomously. ...
متن کامل