Frank L. Hammond

Wearable soft robotic device for post-stroke shoulder rehabilitation: Identifying misalignments

Stroke is the leading cause of long-term disability in the United States, affecting over 795,000 people annually. In order to regain motor function of the upper body, patients are usually treated by regular sessions with a dedicated physical therapist. A cost-effective wearable upper body orthotics system that can be used at home to empower both the patients and physical therapists is described. The system is composed of a thin, compliant, lightweight, cost-effective soft orthotic device with an integrated cable actuation system that is worn over the upper body, an embedded limb position sensing system, an electric actuator package and controller. The proposed device is robust to misalignments that may occur during actuation of the compliant brace or when putting on the system. Through simulations and experimental evaluation, it was demonstrated i) that the soft orthotic cable-driven shoulder brace can be successfully actuated without the production of off-axis torques in the presence of misalignments and ii) that the proposed model can identify linear and angular misalignments online.

Soft tactile sensor arrays for micromanipulation

Micromanipulation methods used for complicated tasks such as microrobot assembly and microvascular surgery often lack the force reflection and contact localization capability necessary to achieve robust grasps of micro-scale objects without applying excessive forces. This absence of haptic feedback is especially prohibitive in cases where visual evidence of force application, such as object surface deformation, is imperceptible and where unstructured, dynamically changing environments require force sensing and modulation for safe, atraumatic object manipulation. This paper describes the design, fabrication, and experimental validation of a soft tactile sensor array for sub-millimeter contact localization and contact force measurement during micromanipulation. The geometry and placement of conductive liquid embedded channels within the sensor array are optimized to provide adequate sensitivity for representative micro-manipulation tasks. Mechanical testing of the sensor demonstrates a sensitivity of less than 50mN and contact localization resolution on the order of 100s of microns.

The effect of implant size and device keel on vertebral compression properties in lumbar total disc replacement.

BACKGROUND CONTEXT Vertebral end plate support is necessary for successful lumbar total disc replacement (TDR) surgery. Failure to achieve anterior column support as a result of lumbar TDR device undersizing could lead to implant subsidence and fracture. PURPOSE The purpose of the study was to examine the compressive biomechanical behavior of the vertebral end plate with varying sizes of disc replacement implants. STUDY DESIGN The study design comprises a biomechanical investigation using a human cadaveric lumbar spine model. METHODS Fifty-six vertebrae with intact posterior elements were prepared from 13 fresh frozen lumbar spines. Peripheral quantitative computed tomography was performed to assess regional bone density. Vertebrae were potted and subjected to nondestructive compression testing with a small, medium, and large custom-made implants with the footplate geometry of the ProDisc-L TDR (Synthes Spine, West Chester, PA, USA) system and having no keel. Failure testing was performed using the ProDisc-L implant with an intact keel. Pressure sensor film was used to assess contact pressure and distribution. RESULTS There was a linear correlation between percent coverage of the end plate and implant-end plate stiffness (p=.0001) and an inverse correlation with displacement (p=.01). The difference in implant-end plate stiffness between small-medium, medium-large, and small-large implants was 10.5% (p=.03), 10.2% (p=.02), and 19.6% (p<.0001), respectively. Failure analysis revealed similar trends for implant sizing, but only bone density was found to significantly correlate with failure properties (r=0.76, p<.0001). There was a significant reduction in implant-end plate stiffness of 18% when the keel was intact compared to without the keel (range 6-27%, p=.0008). Pressure film analysis revealed that the implant was loaded peripherally and did not have central contact during nondestructive loading. There was a trend toward greater contact pressure with the small implant when compared with the medium implant (p=.06) and the large implant (p=.06). CONCLUSIONS Although larger implants reduce end plate displacement, increase apparent implant-end plate stiffness, increase the implant-end plate contact area, and decrease the peak contact pressures, low bone density reduces failure properties. The keel introduces a reduction in stiffness to the implant-end plate interface, the clinical significance of which is currently unknown.

Soft Tactile Sensor Arrays for Force Feedback in Micromanipulation

This paper describes the design, fabrication, and experimental validation of a soft tactile sensor array for submillimeter contact localization and contact force measurement in micromanipulation. The geometry and placement of conductive liquid microchannels embedded within the elastic sensor body are optimized to provide high sensitivity for representative micromanipulation tasks and to overcome functional limitations seen in previous soft tactile sensor research. Mechanical testing of the numerically optimized sensor prototype demonstrates sensitivity to normal contact forces of and submillimeter contact localization resolution. Tactile sensing experiments demonstrate the ability to infer the abstract geometries and motions of objects imparting force on the sensor surface by analyzing microchannel deformation patterns.

Toward a modular soft sensor-embedded glove for human hand motion and tactile pressure measurement

The ability to measure human hand motions and interaction forces is critical to improving our understanding of manual gesturing and grasp mechanics. This knowledge serves as a basis for developing better tools for human skill training and rehabilitation, exploring more effective methods of designing and controlling robotic hands, and creating more sophisticated human-computer interaction devices which use complex hand motions as control inputs. This paper presents work on the design, fabrication, and experimental validation of a soft sensor-embedded glove which measures both hand motion and contact pressures during human gesturing and manipulation tasks. We design an array of liquid-metal embedded elastomer sensors to measure up to hundreds of Newtons of interaction forces across the human palm during manipulation tasks and to measure skin strains across phalangeal and carpal joints for joint motion tracking. The elastomeric sensors provide the mechanical compliance necessary to accommodate anatomical variations and permit a normal range of hand motion. We explore methods of assembling this soft sensor glove from modular, individually fabricated pressure and strain sensors and develop design guidelines for their mechanical integration. Experimental validation of a soft finger glove prototype demonstrates the sensitivity range of the designed sensors and the mechanical robustness of the proposed assembly method, and provides a basis for the production of a complete soft sensor glove from inexpensive modular sensor components.

Towards a design optimization method for reducing the mechanical complexity of underactuated robotic hands

Underactuated compliant robotic hands exploit passive mechanics and joint coupling to reduce the number of actuators required to achieve grasp robustness in unstructured environments. Reduced actuation requirements generally serve to decrease design cost and improve grasp planning efficiency, but overzealous simplification of an actuation topology, coupled with insufficient tuning of mechanical compliance and hand kinematics, can adversely affect grasp quality and adaptability. This paper presents a computational framework for reducing the mechanical complexity of robotic hand actuation topologies without significantly decreasing grasp robustness. Open-source grasp planning software and well-established grasp quality metrics are used to simulate a fully-actuated, 24 DOF anthropomorphic robotic hand grasping a set of daily living objects. DOFs are systematically demoted or removed from the hand actuation topology according to their contribution to grasp quality. The resulting actuation topology contained 22% fewer DOFs, 51% less aggregate joint motion, and required 82% less grasp planning time than the fully-actuated design, but decreased average grasp quality by only 11%.

Design considerations for an active soft orthotic system for shoulder rehabilitation

Strokes affect over 750,000 people annually in the United States. This significant and disabling condition can result in paralysis that must be treated by regular sessions with a dedicated physical therapist in order to regain motor function. However, the use of therapists is expensive, in high demand, and requires patient travel to a rehabilitation clinic. We propose an inexpensive and wearable upper body orthotics system that can be used at home to provide the same level of rehabilitation as the current physical therapy standard of care. The system is composed of a soft orthotic device with an integrated cable actuation system that is worn over the upper body, a limb position sensing system, and an actuator package. This paper presents initial design considerations and the evaluation of a proof of concept system for shoulder joint rehabilitation. Through simulations and experimental evaluation, the system is shown to be adjustable, easily wearable, and adaptable to misalignment and anatomical variations. Insights provided by these initial studies will inform the development of a complete upper body orthotic system.

Morphological design optimization of kinematically redundant manipulators using weighted isotropy measures

Kinematically redundant manipulators are coveted for their ability to perform more complex and a greater variety of tasks than their non-redundant counterparts. This increased utility demands that manipulator designs be carefully optimized to achieve the kinematic dexterity required to perform their numerous intended tasks. The optimization of redundant manipulator designs to improve isotropy has been studied at great length, but a vast majority of the work done focuses on planar manipulation tasks and workspaces that, unlike many modern manufacturing environments, offer few or no physical impediments to motion. In this paper we investigate the incorporation of secondary manipulation goals, in particular obstacle avoidance, into the calculation of kinematic isotropy measures. We will use these weighted isotropy measures as a performance metric for redundant manipulators working in obstacle-laden workspaces, and employ the metric as part of an objective function for a global search design optimization problem. The effectiveness of the weighted isotropy design optimization will be demonstrated by increasing the global dexterity of a sub-optimal seven degree-of-freedom manipulator design used for pick-and-place tasks within a small, enclosed workspace.

Dexterous high-precision robotic wrist for micromanipulation

This paper describes the design and experimental validation of a three degree-of-freedom (DOF) robotic wrist which enables high-precision, anthropomorphic motion suitable for both teleoperative and automated robotic micromanipulation tasks. The proposed parallel-platform based robotic wrist improves upon previous wrist designs by combining the mechanical stiffness and precision of conventional parallel-platform manipulators (PPMs) with the larger workspaces and more dexterous motion of serial chain manipulators (SCMs). This robotic wrist also includes a non-backdrivable actuation mechanism, a continuous tool rotation DOF which allows for non-anthropomorphic twisting motions necessary for drilling and screw mating, and a coaxial channel through which wires and tubes can pass. A dexterous wrist prototype demonstrates an angular motion resolution of 0.1° and a motion bandwidth of 3.0Hz over a motion range greater than that of the human wrist.

Improvement of redundant manipulator task agility using multiobjective weighted isotropy-based placement optimization

The measurement and improvement of task agility, the ability of a manipulator to effectively handle multiple task types, has become increasingly important in the manufacturing industry as efforts are made to design more versatile and efficient factories. Task agility is crucial to the performance of kinematically redundant manipulators in complex manufacturing workspaces that involve several physical motion impediments and conservative dynamic actuation constraints, and which are used to complete a wide variety of manufacturing operations. Manipulator morphology and workspace layout optimizations offer robust, long term solutions to the problem of improving task agility, but the cost of redevelopment or redeployment is typically prohibitive. Manipulator placement optimization is a more economically feasible and practical solution that is amenable to short term implementation. In this paper we propose the use of a multiobjective weighted isotropy measure as a task agility metric and submit it as the basis for optimizing the placement of redundant manipulators in a complex, multitask workspaces. We describe the formulation of this measure, how it factors motion impedance into kinematic dexterity, and the advantages of this measure over previous task agility measurement methods. We demonstrate its efficacy in improving task agility by optimizing manipulator base placement to achieve collision-free motion and reduced maximum torques across an entire set of disparate manipulation tasks.