Jaydeep Roy

Adaptive force control of position/velocity controlled robots: theory and experiment

This paper addresses the problem of achieving exact dynamic force control with manipulators possessing low-level position and/or velocity controllers typically employed in industrial robot arms. Previously reported approaches and experimental results are reviewed. A new adaptive force control algorithm for velocity/position controlled robot arms in contact with surfaces of unknown linear compliance is reported. The controller provably guarantees global asymptotic convergence of force trajectory tracking errors to zero when the robot is under exact or asymptotically exact inner loop velocity control. An additional result which guarantees arbitrarily small force errors for bounded inner loop velocity tracking errors is presented. Comparative experiments show the new adaptive velocity (position) based controller and its nonadaptive counterpart to provide performance superior to that of previously reported position-based force controllers.

Robot-Assisted Stapedotomy: Micropick Fenestration of the Stapes Footplate

OBJECTIVE: Micropick fenestration of the stapes footplate, a difficult step in stapedotomy, was selected for trials evaluating the potential for robotic assistance (RA) to improve clinical measures of surgical performance. STUDY DESIGN: In a surgical model of stapedotomy, we measured accuracy of fenestration to a desired point location and force applied to the stapes footplate. Performance variables were measured for 3 experienced and 3 less–experienced surgeons. RESULTS: RA significantly reduced the maximum force applied to the stapes footplate. For fenestration targeting, RA significantly improved accuracy for less–experienced surgeons and significantly worsened targeting for more-experienced surgeons. CONCLUSIONS: RA significantly improves performance for micropick fenestration in a surgical model of stapedotomy. For certain tasks, RA differentially affects performance for users of different experience levels. CLINICAL SIGNIFICANCE: These are the first results showing quantitative improvements in performance during simulated ear surgery using RA and differential effects of RA on performance for users of different experience levels.

Performance Evaluation of a Cooperative Manipulation Microsurgical Assistant Robot Applied to Stapedotomy

This paper reports the development of a full-scale instrumented model of the human ear that permits quantitative evaluation of the utility of a microsurgical assistant robot in the surgical procedure of stapedotomy.

Task performance in stapedotomy: Comparison between surgeons of different experience levels

OBJECTIVES: Two steps in stapedotomy are particularly challenging: (1) micropick fenestration of the stapes footplate (SF) and (2) crimping of the stapes prosthesis (SP) to the incus. We conducted trials to determine if experience correlates with differences in performance for these tasks. METHODS: In a surgical model of stapedotomy, performance was measured for 3 experienced and 3 novice surgeons. For fenestration, we measured ability to target the fenestration and force applied to the SF. For crimping, we measured crimp quality, movement of the SP during crimping, and force applied to the SF. RESULTS: Experienced surgeons demonstrated significantly better ability to target the fenestration and, during crimping, caused less SP movement and a significantly lower rate of SP dislodgment. CONCLUSIONS: Clear differences in task performance are measurable between more and less experienced surgeons during critical steps of stapedotomy. CLINICAL SIGNIFICANCE: The observed differences in task performance may contribute to an understanding of maneuvers that increase the risk of inadequate prosthesis placement and cochlear trauma—factors likely responsible for variable hearing results with strapedotomy.

Haptic feedback augmentation through position based adaptive force scaling: theory and experiment

This paper addresses the development, implementation, and evaluation of a novel application of robot force control called position based force scaling. We report the first results in the implementation of asymptotically exact force scaling, as well as the first results in on-line environment compliance adaptation while performing the task of force scaling. One previously reported and two new (one adaptive and one non-adaptive) force scaling algorithms are reviewed. Extensive comparative experiments quantify the performance of all three algorithms and show that the new adaptive force scaling algorithm significantly outperforms both its non-adaptive counterpart as well as the previously reported set-point-regulator based force scaling algorithm over a wide range of user applied input force trajectories and force scale factors.

Comparative structural analysis of 2-DOF semi-direct-drive linkages for robot arms

Reports the first direct structural comparison of a family of two-degrees-of-freedom semi-direct-drive linkages for robot arms. One novel and three previously reported linkage designs were optimized and compared using finite-element analysis. The results quantify structural properties of the four optimized designs under variations in joint position and payload. Several frequently overlooked issues essential for fair comparison of disparate linkages are identified and addressed.

Multi-source inverse-geometry CT: From system concept to research prototype

Third-generation CT architectures are approaching fundamental limits. Dose-efficiency is limited by finite detector efficiency and by limited control over the X-ray flux spatial profile. Increasing the volumetric coverage comes with increased scattered radiation, cone-beam artifacts, Heel effect, wasted dose and cost. Spatial resolution is limited by focal spot size and detector cell size. Temporal resolution is limited by mechanical constraints, and alternative geometries such as electron-beam CT and dual-source CT come with severe tradeoffs in terms of image quality, dose-efficiency and complexity. The concept of multi-source inverse-geometry CT (IGCT) breaks through several of the above limitations [1-3], promising a low-dose high image quality volumetric CT architecture. In this paper, we present recent progress with the design and integration efforts of the first gantry-based multi-source CT scanner.

Structural design, analysis, and performance evaluation of a new semi-direct drive robot arm: theory and experiment

This paper reports the mechanical design, structural analysis, and experimental verification of a new high-performance semi-direct drive robot arm. A design-optimization methodology employing finite element analysis (FEA) is reviewed, and a resulting arm design is reported. FEA simulations of the final design predict high structural vibration frequencies throughout the arms workspace. Extensive structural vibration experiments with the completed manipulator confirm the predicted structural vibration characteristics throughout the arms workspace. Position-tracking experiments show that the manipulator accurately tracks fast time-varying reference trajectories-peak tip velocities greater than 6 m/s and peak tip accelerations greater than 7 g.

Structural design optimization and comparative analysis of a new high-performance robot arm via finite element analysis

This paper reports the structural design of a new high-performance robot arm. Design objectives for the new arm include large (1-2m) workspace, low weight, 5 kg payload capacity, high stiffness, high structural vibration frequencies, precise joint-level torque control, a total of three degrees-of-freedom, and mechanical simplicity. A comparative analysis is reported for four very different two degree-of-freedom linkage candidates using the finite element method.

Haptic Feedback Enhancement Through Adaptive Force Scaling: Theory and Experiment

We report the development, implementation, and evaluation of a novel application of robot force control called position based force scaling. Force scaling employs position based force control algorithms to augment human haptic feedback during human-robot co-manipulation tasks. We report the first results in the implementation of asymptotically exact force scaling, as well as the first results in on-line environment compliance adaptation while performing the task of force scaling. One previously reported and two new (one adaptive and one non-adaptive) force scaling algorithms are reviewed. Comparative experiments are reported which quantify the performance of all three algorithms. The data show that the new adaptive force scaling algorithm significantly outperforms both its non-adaptive counterpart as well as