Modeling and Planning Manipulation in Dynamic Environments

Schmitt, Philipp; Witnshofer, Florian; Wurm, Kai M.; Wichert, Georg von; Burgard, Wolfram

doi:10.1109/icra.2019.8793824

Cited by 27 publications

(8 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Schmitt et al [48] propose an approach where two robots manipulate an object in a dynamic environment. The output of the algorithm is a sequence of controllers rather than a sequence of paths.…”

Section: B Manipulation Planningmentioning

confidence: 99%

Prehensile Manipulation Planning: Modeling, Algorithms and Implementation

Lamiraux

Mirabel

2022

IEEE Trans. Robot.

View full text Add to dashboard Cite

This paper presents a software platform tailored for prehensile manipulation planning named Humanoid Path Planner. The platform implements an original way of modeling manipulation planning through a constraint graph that represents the numerical constraints that define the manipulation problem. We propose an extension of the RRT algorithm to manipulation planning that is able to solve a large variety of problems. We provide replicable experimental results via a docker image that readers may download to run the experimental results by themselves.

show abstract

Section: B Manipulation Planningmentioning

confidence: 99%

Prehensile Manipulation Planning: Modeling, Algorithms and Implementation

Lamiraux

Mirabel

2022

IEEE Trans. Robot.

View full text Add to dashboard Cite

show abstract

“…4) Motion Generation: When obstacles are present, reaching a moving target requires some trajectory planning (such as RRT [24], PRM [25]) or trajectory optimization methods (such as CHOMP [26], STOMP [27]) that are able to generate collision-free paths for the arm. Recent works [28], [29] presented an approach to generate a sequence of constraint-based controllers to reactively execute a plan while respecting specified constraints like collision avoidance. Our work is more similar to works that generate arm motion from a library of stored arm motions [30], [31], [32].…”

Section: Dynamic Graspingmentioning

confidence: 99%

Dynamic Grasping with Reachability and Motion Awareness

Akinola

Song

et al. 2021

2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

Grasping in dynamic environments presents a unique set of challenges. A stable and reachable grasp can become unreachable and unstable as the target object moves, motion planning needs to be adaptive and in real time, the delay in computation makes prediction necessary. In this paper, we present a dynamic grasping framework that is reachabilityaware and motion-aware. Specifically, we model the reachability space of the robot using a signed distance field which enables us to quickly screen unreachable grasps. Also, we train a neural network to predict the grasp quality conditioned on the current motion of the target. Using these as ranking functions, we quickly filter a large grasp database to a few grasps in real time. In addition, we present a seeding approach for arm motion generation that utilizes solution from previous time step. This quickly generates a new arm trajectory that is close to the previous plan and prevents fluctuation. We implement a recurrent neural network (RNN) for modelling and predicting the object motion. Our extensive experiments demonstrate the importance of each of these components and we validate our pipeline on a real robot.

show abstract

“…The paper presents impressive experimental results with a quadruped robot equipped with a lidar and avoiding unexpected obstacles. [21] proposes a method to plan a sequence of controllers that move a system composed of two robots and one object from an initial state to a goal state in a dynamic environment. The algorithm performs a random exploration of the configuration space, using simulated controllers as a steering method.…”

Section: A Manipulation Planningmentioning

confidence: 99%