Planning motions to grasp an object in cluttered and uncertain environments is a challenging task, particularly when collision-free trajectory does not exist objects obstructing the way are required be carefully grasped moved out. This letter takes different approach proposes address this problem by using randomized physics-based motion planner that permits robot–object object–object interactions. The main idea avoid explicit high-level reasoning of task providing with physics engine...

This paper presents the software tool used at Institute of Industrial and Control Engineering (IOC-UPC) for teaching research in robot motion planning. The allows to cope with problems one or more robots, being a generic defined as kinematic tree mobile base, i.e. can plan simulate from simple two degrees freedom free-flying robots multi-robot scenarios manipulators equipped anthropomorphic hands. main core planners is provided by Open Motion Planning Library (OMPL). Different basic be...

Autonomous indoor service robots are supposed to accomplish tasks, like serve a cup, which involve manipulation actions. Particularly, for complex tasks subject geometric constraints, spatial information and rich semantic knowledge about objects, types, functionality required, together with the way in these objects can be manipulated. In this line, paper presents an ontological-based reasoning framework called Perception Manipulation Knowledge (PMK) that includes: (1) modeling of environment...

This paper presents a planning procedure that allows an anthropomorphic dual-arm robotic system to perform manipulation task in natural human-like way by using demonstrated human movements. The key idea of the proposal is convert trajectories into attractive potential fields defined over configuration space and then use RRT*-based algorithm minimizes path-cost function designed bias tree growth toward human-demonstrated configurations. description proposed approach as well results from...

Potential-field approaches based on harmonic functions have good path planning properties, although the explicit knowledge of robot's Configuration Space is required. To overcome this drawback, a combination with random sampling scheme proposed. Harmonic are computed over 2 d-tree decomposition d-dimensional that obtained probabilistic cell (sampling and classification). Cell biased towards more promising regions by using function values. classification performed evaluating set...

The paper deals with the problem of planning movements dual-arm anthropomorphic systems, aim reducing computational cost and making look as human-like possible. key idea proposal is search synergies system in order to use them reduce dimension space while preserving appearance. This was already developed successfully used plan robotic hands, thus extension a attractive. presents description proposed approach well real experimental results that encourage doing further research this line.

Traditionally, grasp and arm motion planning are considered as separate tasks. This paper presents an integrated approach that only requires the initial configuration of robotic pose target object to simultaneously plan a good hand trajectory object. The planner exploits concept independent contact regions look for best possible grasp. goal poses end effector obtained using two different methods: one biases sampling towards favorable principal component analysis, another considers...

Teleoperating a robotic hand with the aid of sensorized glove presents some particular problems.One such problem is due to kinematic differences between human and hand, which do not allow simple direct mapping sensor readings from hand.This addressed different types mapping, but none them general use.This paper proposes two new mappings within existing types, as well hybrid that combines best features these types.This allows intuitive free space movements (where gesture more important than...

Robotic manipulation involves actions where contacts occur between the robot and objects. In this scope, availability of physics-based engines allows motion planners to comprise dynamics rigid bodies, which is necessary for planning type actions. However, computationally intensive due high dimensionality state space need work with a low integration step find accurate solutions. On other hand, change environment conditions further motions. To cope issue, representation using ontologies...

This paper discusses motion planning for anthropomorphic dual-arm robots. It introduces a measure of the similarity movements needed to solve two given tasks. Planning using this select proper arm synergies task improves performance and resulting plan.

For everyday manipulation tasks, the combination of task and motion planning is required regarding need providing set possible subtasks which have to be done how perform them. Since many alternative plans may exist, determination their feasibility identification best one a great challenge in robotics. To address this, this paper proposes: a) version GraphPlan (one current approaches planning) that has been modified use ontological knowledge allow retrieval all plans; b) physics-based...

Robotic systems composed of several mobile robots moving in human environments pose problems at perception, planning and control levels. In these environments, there may be obstacles obstructing the paths, which can remove by pushing or pulling them. At level, therefore, an efficient combination task motion is required. Even more if we assume a cooperative system collaborate with each other e.g. together heavy obstacle one robot clearing way to another one. this paper, cope problem proposing...

The automatic synthesis of a plan to perform robotized assembly task from high-level description the product be assembled is difficult issue which involves planning and planning. This paper summarizes problems that an autonomous robotic system must tackle both regarding reviews approaches use Petri nets as formalism develop corresponding planners. need for integration made evident, covering are described.

This article presents a graphical task-level robot-programming tool for polishing parts held by the robot gripper. The proposed approach allows user-friendly manner to specify curves over CAD model of part. user also specifies width abrasive bands be used and pressure exerted. GUI aids in verifying that whole surface is correctly polished. Once task specification done, task-planning module provides algorithms guarantee time-optimum sequence trajectories perform part, avoiding collisions with...

In order to solve mobile manipulation problems, the efficient combination of task and motion planning is usually required. Moreover, incorporation physics-based information has recently been taken into account in plan tasks a more realistic way. present paper, framework proposed based on modified version Fast-Forward planner that guided by knowledge. The proposal uses knowledge for reasoning symbolic literals (both offline online modes) taking geometric evaluate applicability as well...

Abstract PhD programs and graduate studies in robotics usually include motion planning among its main subjects. Students that focus their research this subject find themselves trapped the necessity of programming an environment where to test validate theoretic contributions. The robot is a big challenge. It requires on one hand good skills involving use software development tools, paradigms, or knowledge computational complexity efficiency issues. On other it coping with different related...