Research on robotic manipulation has primarily focused on grasping rigid objects using a single manipulator. It is however evident that in order to be truly pervasive, service robots will need to handle deformable objects, possibly with two arms. In this paper we tackle the problem of using cooperative manipulators to perform towel folding tasks. Differently from other approaches, our method executes what we call a momentum fold -a swinging motion that exploits the dynamics of the object being manipulated. We propose a new learning algorithm that combines imitation and reinforcement learning. Human demonstrations are used to reduce the search space of the reinforcement learning algorithm, which then quickly converges to its final solution. The strengths of the algorithm come from its efficient processing, fast learning capabilities, absence of a deformable object model, and applicability to other problems exhibiting temporally incoherent parameter spaces. A wide range of experiments were performed on a robotic platform, demonstrating the algorithm's capability and practicality.
Abstract-We consider the problem of searching for an unknown number of static targets inside an assigned area. The search problem is tackled using Probabilisitic Quadtrees (PQ), a data structure we recently introduced. Probabilistic quadtrees allow for a variable resolution representation and naturally induce a search problem where the searcher needs to choose not only where to sense, but also the sensing resolution. Through a Bayesian approach accommodating faulty sensors returning both false positives and missed detections, a posterior distribution about the location of the targets is propagated during the search effort. In this paper we extend our previous findings by considering the problem of searching for an unknown number of targets. Moreover, we substitute our formerly used heuristic with an approach based on information gain and expected costs. Finally, we provide some convergence results showing that in the worst case our model provides the same results as uniform grids, thus guaranteeing that the representation we propose gracefully degrades towards a known model. Extensive simulation results substantiate the properties of the method we propose, and we also show that our variable resolution method outperforms traditional methods based on uniform resolution grids.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.