Divide-and-conquer manipulation planning by lazily searching a weighted two-layer manipulation graph

This paper presents a manipulation planning algorithm for robots to reorient objects. It automatically finds a sequence of robot motion that manipulates and prepares an object for specific tasks. Examples of the preparatory manipulation planning problems include reorienting an electric drill to cut holes, reorienting workpieces for assembly, and reorienting cargo for packing, etc. The proposed algorithm could plan single and dual arm manipulation sequences to solve the problems. The mechanism under the planner is a regrasp graph which encodes grasp configurations and object poses. The algorithms search the graph to find a sequence of robot motion to reorient objects. The planner is able to plan both single and dual arm manipulation. It could also automatically determine whether to use a single arm, dual arms, or their combinations to finish given tasks. The planner is examined by various humanoid robots like Nextage, HRP2Kai, HRP5P, etc., using both simulation and real-world experiments.

show abstract

“…For configuration spaces connected by narrow passages, DD-RRT is a bit slow. Methods like [30] are preferable.…”

Section: A Single-arm Planningmentioning

confidence: 99%

Preparatory Manipulation Planning Using Automatically Determined Single and Dual Arm

Wan

Harada

Kanehiro

2020

IEEE Trans. Ind. Inf.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Cho, Kim, and Song (2003) used discretization to build a look-up table of grasps and placements then searched for a solution in the table. Wan et al (2015), Wan and Harada (2015a), and Wan and Harada (2015b) constructed a two-layer regrasp graph where the rst layer indicated how di erent placement classes could be connected to one another, and the second layer indicated how di erent grasps could be connected to one another.…”

Section: Unimanual Manipulation Planningmentioning

confidence: 99%

Planning algorithms for complex manipulation tasks

Lertkultanon¹

View full text Add to dashboard Cite

Solving manipulation tasks requires planning not only robot motions but also various interaction such as grasps (robot-object) and placements (object-environment). This indispensable interaction imparts extra complexity to the problems such that solving complex manipulation tasks, which require a number of regrasping operations, remains elusive. In this thesis, we advance the state of the art by presenting novel unimanual and bimanual manipulation planning algorithms capable of planning manipulation motions with multiple regrasping. First, we introduce a unimanual manipulation planner that explores the composite con guration space e ciently and systematically, thanks to the guidance of the novel high-level grasp-placement graph. Unlike existing methods, the graph construction does not require heavy preprocessing and is speci c to only the gripper and the manipulated object. Next, we present two bimanual manipulation planners. The rst one addresses speci c, yet challenging, cases when bimanual grasps remain the same throughout. With the novel characterization of con guration space with closed-chain constraints, the proposed planner can plan motions across di erent closed-chain connected components. The second one addresses more general cases when the object can be moved only when grasped by both robots. We present a planner with certi ed completeness property, which guarantees that when a certi cate is available for a given object and environment, the planner will nd a solution to any bimanual manipulation query whenever one exists. The hardware experiment demonstrates the planner's capability and is, to the best of our knowledge, the rst to illustrate such regrasping capability, solving complex bimanual manipulation task on an actual system. Furthermore, we also present two improvements to motion planning, which indeed is a crucial component in any manipulation planning algorithm. The rst improvement is an algorithm for generating time-optimal second-order trajectories subject to velocity, acceleration, and minimum-switch-time constraints. The latter constraint helps prevent concentrated acceleration switching in trajectories. The second improvement is a new bidirectional motion planner called AVP-BiRRT. The integration of the Admissible Velocity Propagation (AVP) algorithm, which enables a geometric path planner to nd dynamically feasible paths, into a bidirectional planner is made possible by our newly proposed extension, AVP-Backward. 4First and foremost, I would like to thank my supervisor Quang-Cuong Pham. It is he who steers me in the right direction throughout my study. I am thankful for his support, his care, and his time he has given to me, which are, I truly believe, among the best a PhD student could hope for. I appreciate how his perpetual energy and enthusiasm keep pushing me forward. We have had countless fruitful discussions and I have learned a lot from him.I am grateful to the good friends I have here in CRI group-Huy Nguyen, Francisco Suárez-Ruiz, Ahmad Bin Anwar, Tien Hung Pham, Aditya Kapoor, ...

show abstract

Preparatory Manipulation Planning using Automatically Determined Single and Dual Arms

Wan¹,

Harada²,

Kanehiro³

2018

Preprint

View full text Add to dashboard Cite

Divide-and-conquer manipulation planning by lazily searching a weighted two-layer manipulation graph

Cited by 3 publications

References 31 publications

Preparatory Manipulation Planning Using Automatically Determined Single and Dual Arm

Preparatory Manipulation Planning Using Automatically Determined Single and Dual Arm

Planning algorithms for complex manipulation tasks

Preparatory Manipulation Planning using Automatically Determined Single and Dual Arms

Contact Info

Product

Resources

About