A new interaction force decomposition maximizing compensating forces under physical work constraints

Schmidts, Alexander M.; Schneider, Manuel K.; Kühne, Markus; Peer, Angelika

doi:10.1109/icra.2016.7487698

Cited by 4 publications

(19 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although it is in general not possible to achieve full decomposition of interactions into IWs leading to wrenches compensating each other and MWs contributing to the resulting wrench only without compensating parts [38], the idea of a non-squeezing pseudoinverse [35] can be employed in wrench synthesis problem. Different from the statement in [35] that the non-squeezing load distribution is unique, we will show a more general formulation with non-squeezing effect in desired MWs.…”

Section: B Internal Wrench Regulationmentioning

confidence: 99%

Fully Distributed Cooperation for Networked Uncertain Mobile Manipulators

2020

View full text Add to dashboard Cite

To achieve distributed control of a multi-arm system, leaderfollower approach is an option generally associated with the schemes that are devoted to reducing the communication burden while trying to achieve as far as possible. Based on the diagram of impedance dynamics and leader-follower scheme, coordinated motion control for multiple mobile manipulators is employed to achieve cooperative object manipulation [10]. Inspired by a team of people moving a table, the followers can implement similar impedance law as the leader's either by estimating the leader's desired motion [11] or by taking the contact force as the leader's motion intention [12]. This innovation enables the whole system to work under implicit communication. Another interesting work [13] that does not require communication achieves force coordination between leader and follower only by measuring the object's motion as the feedback. However, the assumptions that all followers act passively in the transport task in [10], [11], that desired velocity of the grasped object is constant and available to each agent in [12], that the attachment points of the collectives are centrosymmetric around the center of mass (COM) of the object in [13], act as the primary factors that hind the applications of their works to our case.The idea that successful object transportation in real application is generally strongly related to the robot formation has also been continuously inspiring related works [14], [15]. Along with the rapid advances in graph theory and control philosophy of the multi-agent systems, distributed scheme under explicit communication plays an important role in the formation control of multiple mobile robots [16], [17]. The challenge existing in formation-based transport task for multiple mobile manipulator ensemble lies in the design of a distributed control law to achieve a global behavior in cooperative manner with limited local information and constrained communication [18]. A typical schema [19] employs a set of distributed controller/observer to achieve relative formation of the multi-arm system. Convergent estimation of the collective states by local observer bridges the gap between the local control and the global cooperative behavior, thus a totally distributed cooperation is achieved [20].To further maintain high performance when the mobile robot team executes tightly cooperative task whilst suffering from the inevitable uncertainties of the robot dynamics, adaptive mechanism is introduced into the architecture, based on either impedance control [21] or force/position control [19]. More complicate case in which the dynamic uncertainty and communication constraints (e.g. the jointly communication topology [2]) coexist can be easily tackled by embedding the parameter adaptation to the respective control scheme. Recent representative work [22] employs the robust adaptive control to concurrently address dynamic uncertainties and external disturbances. The dependence of the communication network and the costly force/torque sensor is further e...

show abstract

Section: B Internal Wrench Regulationmentioning

confidence: 99%

Fully Distributed Cooperation for Networked Uncertain Mobile Manipulators

2020

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

“…An important step towards physically plausible wrench decomposition was recently taken by Schmidts et al in [25], by introducing force decomposition constraints motivated by mechanical work. The wrench decomposition solution for two effectors proposed in [26] satisfies the proposed constraints of [25]. Erhart and Hirche recently suggested a different decomposition approach for cooperative object manipulation that also includes the application of torque in [27] and is based on kinematic constraint violation of desired accelerations as presented in [28].…”

Section: Introductionmentioning

confidence: 99%

“…Erhart and Hirche recently suggested a different decomposition approach for cooperative object manipulation that also includes the application of torque in [27] and is based on kinematic constraint violation of desired accelerations as presented in [28]. One of the main findings of their works is the existence of infinite different pseudoinverses of the grasp matrix that specify desired load shares of the effectors, although their computation of internal wrench does not necessarily comply with the constraints of [25].…”

Section: Introductionmentioning

confidence: 99%

“…• an extension of the force constraints proposed by [25] to the application of torque, • a reformulation of the optimization proposed by [25] based on physical plausibility considerations yielding a convex optimization problem, • derivation of analytic solutions for special cases • and wrench measures for analysis of pHRI and pHHI tasks. The result is a physically plausible wrench decomposition into manipulation and internal components for rigid object manipulation.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Physically Plausible Wrench Decomposition for Multieffector Object Manipulation

2018

View full text Add to dashboard Cite

When manipulating an object with multiple effectors such as in multi-digit grasping or multi-agent collaboration, forces and torques (i.e. wrench) applied to the object at different contact points generally do not fully contribute to the resultant object wrench, but partly compensate each other. The current literature, however, lacks a physically plausible decomposition of the applied wrench into its manipulation and internal components. We formulate the wrench decomposition as a convex optimization problem, minimizing the Euclidean norms of manipulation forces and torques. Physical plausibility in the optimization solution is ensured by constraining the internal and manipulation wrench by the applied wrench. We analyze specific cases of 3digit grasping and 2D beam manipulation, and show the applicability of our method to general object manipulation with multiple effectors. The wrench decomposition method is then extended to quantification of measures important in evaluating physical human-human and human-robot interaction tasks. We validate our approach via comparison to the state of the art in simulation and via application to a human-human object transport study.

show abstract

Online Capability Based Task Allocation of Cooperative Manipulators

Patra,

Sinha,

Guha

2024

J Intell Robot Syst

View full text Add to dashboard Cite

The cooperative manipulator group can accomplish complex and heavy payload tasks of object manipulation and transportation compared to a single manipulator. Effective coordination is crucial for cooperative task accomplishments. Multi-manipulator task distribution is highly complex because of the varying dynamic capabilities of the manipulators. We have introduced a novel fastest technique to quantify the dynamic task capability of the cooperative manipulator by scalar quantity and allocate the task accordingly. The scalar quantity determines the capability of applying an external wrench by end effector (EE) in line with the required wrench at the center of mass of the manipulating object. This quantity helps to diminish tracking errors in object manipulations or transportation and actuator saturation avoidance. The task distribution among the members is in proportion to their computed dynamic capability to ensure equal priority to the individual manipulators. The proposed task distribution formulation ensures the minimum magnitude of wrench interaction at the grasp point and the minimum internal wrench build-up in the object. Several physical simulation results assure trajectory tracking performance with the proposed task capability metric. The same metric aids in identifying the least capable manipulator, rearranging members for better performance, and deciding the required number of manipulators in the manipulator group.

show abstract

A new interaction force decomposition maximizing compensating forces under physical work constraints

Cited by 4 publications

References 23 publications

Fully Distributed Cooperation for Networked Uncertain Mobile Manipulators

Fully Distributed Cooperation for Networked Uncertain Mobile Manipulators

Physically Plausible Wrench Decomposition for Multieffector Object Manipulation

Online Capability Based Task Allocation of Cooperative Manipulators

Contact Info

Product

Resources

About