Task Performance Evaluation of Asymmetric Semiautonomous Teleoperation of Mobile Twin-Arm Robotic Manipulators

Abstract

A series of human factors experiments involving maneuvering and grasping tasks are carried out to evaluate the effectiveness of a novel asymmetric semiautonomous teleoperation (AST) control design framework for teleoperation of mobile twin-arm robotic manipulators. Simplified configurations are examined first to explore control strategies for different aspects of such teleoperation tasks. These include teleoperation of a nonholonomic mobile base, telemanipulation of a dual-arm robot, and dual-arm/dual-operator teleoperation task scenarios. In two sets of experiments with a planar nonholonomic mobile base, teleoperation via a 3DOF planar haptic interface with position mapping and force reflection of the nonholonomic constraint decreases task-completion-time (TCT) and reduces unwanted collisions. In dual-arm and dual-operator teleoperation maneuverability experiments, the assignment of decoupled and nonconflicting control frames reduces TCT and unwanted contacts. The use of so-called "soft" constraints via passive semiautonomous control reduces TCT and unwanted block drops in telegrasping experiments with a twin-arm manipulator. A final comprehensive experiment encompassing elements of the simplified configurations demonstrates the effectiveness of AST control framework in dual-operator teleoperation of a twin-arm mobile manipulator.

Investigating multiple human performance measures in teleoperation task: A translation task in a tele‐sandblasting maintenance system

“…Task completion time, or MT, is frequently used in evaluations of human performance because it is a measure that relates directly to production rate (Kaber, Li, Clamann, & Lee, 2012;King et al, 2009;Malysz & Sirouspour, 2013;Suebnukarn, Phatthanasathiankul, Sombatweroje, Rhienmora, & Haddawy, 2009;Wildenbeest, Abbink, Heemskerk, Van Der Helm, & Boessenkool, 2013). Some researchers (Chotiprayanakul et al, 2011;Son et al, 2011) have used velocity of movement or average velocity of movement instead of task completion time.…”

“…Teleoperation systems can simplify tasks by scaling up or down the movement between master controller and slave device and can reduce risk to humans by performing tasks in hazardous or difficult to access areas. For example, teleoperation systems have been incorporated into diverse work contexts, such as structure maintenance (Chotiprayanakul, Liu, & Paul, 2011), nuclear/explosive/toxic material handling and disposal (Kron, et al, 2004), rescue missions (Hirche, Stanczyk, & Buss, 2003), entertainment applications (Malysz & Sirouspour, 2013), and da Vinci surgical systems (Polushin, Takhmar, & Patel, 2015). Although teleoperation systems provide many benefits, there has been evidence that teleoperation systems degrade human performance and task performance to control operational device (Nash, Edwards, Thompson, & Barfield, 2000;Stanney, Mourant, & Kennedy, 1998).…”

High‐gain nonlinear observer‐based impedance control for deformable object cooperative teleoperation with nonlinear contact model

“…Multilateral teleoperation is suitable for the dexterous and complicated tasks. Siciliano and Khatib et al concluded multilateral telerobotics into four cases, that is, the multiple‐master single‐slave systems, single‐master multiple‐slave systems, multiple‐master multiple‐slave (MMMS) systems, and conventional teleoperation setup. This article studies a novel kind of MMMS system.…”

Whole-body multi-modal semi-autonomous teleoperation of mobile manipulator systems

“…2. In contrast to the abundance of results on the teleoperation of manipulators, results on the teleoperation of mobile robots or mobile manipulator are relatively rare, e.g., mobile robots [9], aerial robots [10], [11], [12], [13], mobile manipulators [14], [15]. Yet, all those results do not explicitly aim to minimize the dexterity loss and the information loss of telerobotic systems as we attempt to here, with all of them either utilizing only kinesthetic interfaces (with limited-DOF) instead of the operator's whole-body dexterity (e.g., [9], [10], [11], [12], [14], [15]); or focusing only to kinesthetic haptic feedback (e.g., [9], [10], [11], [12], [15]) or even completely missing any haptic feedback (e.g., [13], [16]) in contrast to our visuo-haptic-vestibular feedback.…”