Push-manipulation of complex passive mobile objects using experimentally acquired motion models

Abstract

In a realistic mobile push-manipulation scenario, it becomes non-trivial and infeasible to build analytical models that will capture the complexity of the interactions between the environment, each of the objects, and the robot as the variety of objects to be manipulated increases. We present an experience-based push-manipulation approach that enables the robot to acquire experimental models regarding how pushable real world objects with complex 3D structures move in response to various pushing actions. These experimentally acquired models can then be used either (1) for trying to track a collision-free guideline path generated for the object by reiterating pushing actions that result in the best locally-matching object trajectories until the goal is reached, or (2) as building blocks for constructing achievable push plans via a Rapidly-exploring Random Trees variant planning algorithm we contribute and executing them by reiterating the corresponding trajectories. We extensively experiment with these two methods in a 3D simulation environment and demonstrate the superiority of the achievable planning and execution concept through safe and successful pushmanipulation of a variety of passively mobile pushable objects. Additionally, our preliminary tests in a real world scenario, where the robot is asked to arrange a set of chairs around a table through achievable pushmanipulation, also show promising results despite the increased perception and action uncertainty, and verify the validity of our contributed method.

Learning modular and transferable forward models of the motions of push manipulated objects

“…Most such models are of qualitative effects (Montesano et al 2008;Moldovan et al 2012;Hermans et al 2011;Fitzpatrick et al 2003;Ridge et al 2010;Kroemer and Peters 2014), although metrically precise models have been learned (Meriçli et al 2014;Scholz and Stilman 2010). These learn action-effect correlations.…”

“…Action outcomes were learned for various tools in a modular fashion, but without transfer learning. In both (Meriçli et al 2014) and Scholz and Stilman (2010) the metric planar motion of pushed objects on the plane is learned, and the learning is modularized by object, as here. In each case a small number of discrete pushing actions is tried, and motion models are planar, rather than full rigid body transformations.…”

Learning non-holonomic object models for mobile manipulation

“…Finally, there are several results in object pushing without explicit physical knowledge [13,14], however these approaches are restricted to holonomic objects. We are interested in tasks in human environments, which can frequently include objects with wheels and hinges.…”

Nonprehensile whole arm rearrangement planning on physics manifolds

“…More recent solutions to the rearrangement problem use physics based actions, such as pushing, to achieve more efficient solutions [5,12,25]. For example, the robot may use one arm to push each item to a location where it can be grasped, lifted and stored using the other arm.…”