Haptic-Based Shared-Control Methods for a Dual-Arm System

Abstract

We propose novel haptic guidance methods for a dual-arm telerobotic manipulation system, which are able to deal with several different constraints, such as collisions, joint limits, and singularities. We combine the haptic guidance with shared-control algorithms for autonomous orientation control and collision avoidance meant to further simplify the execution of grasping tasks. The stability of the overall system in various control modalities is presented and analyzed via passivity arguments. In addition, a human subject study is carried out to assess the effectiveness and applicability of the proposed control approaches both in simulated and real scenarios. Results show that the proposed haptic-enabled shared-control methods significantly improve the performance of grasping tasks with respect to the use of classic teleoperation with neither haptic guidance nor shared control.

Haptic-guided shared control for needle grasping optimization in minimally invasive robotic surgery

“…To successfully perform a suturing task without hands-off movements, we guide the human operator towards a grasping pose which yields no singularities or joint limits during the post-grasping suturing trajectory. In contrast to previous studies [1], [14], here we know the geometrical shape of the Fig. 3.…”

“…The partial derivatives of (8) and (9) w.r.t. q s can be easily computed analytically (see [1], [14]) and then plugged in (17) to evaluate (14), thus finding the optimal grasping parameter vector z * according to (13). The optimal Cartesian pose for the PSM x g,d can be easily calculated from the optimal grasping parameter vector z * given the needle kinematics (5) and its global pose (1).…”

“…On the other hand, an operator can be successfully guided towards an optimal grasping configuration. To this end, haptic guidance has been used to avoid constraints during g 0 [14], [15] and g 2 [1]. These works reported significant performance improvement compared to direct teleoperation.…”

“…We use a cost function which is comprised of two terms: (i) joint limits and (ii) task-oriented velocity (TOV) manipulability 1 . In contrast to previous works [1], [14], we assume the shape of the object to be grasped, i.e. the needle, is known.…”

Caring About the Human Operator: Haptic Shared Control for Enhanced User Comfort in Robotic Telemanipulation

“…Shared control algorithms enable to share the available degrees of freedom of the robotic system between the operator and an autonomous controller. Applications include mobile robotics [1], robot-assisted surgery [2], and assistive robotics [3], [4]. Possible implementations of this approach can be obtained using variable admittance control [5], virtual non-holonomic constraints [6]- [8], and online adaptation of the level of support [9].…”