Search

Thesis: S.B., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2018., Cataloged from PDF version of thesis., Includes bibliographical references (pages 39-40)., Stable human locomotion may be described as a non-linear limit cycle oscillator. This claim has been supported through the observation of dynamic entrainment and phase-locking to external mechanical perturbations applied at the ankle. Simple models have been developed in attempts to understand these behaviors, but have been unsuccessful at replicating experimental studies. In this manuscript, an energy-based controller was implemented on a single degree-of-freedom model, adjusting its leading leg angle at heel strike and consequently the energy dissipation of the model. Stochasticity was applied to the controller to simulate the variability which has been observed and quantified in walking. The results indicate that energy control may be responsible for entrainment in human walking, but a revised model may be required to match the experimental coefficients of variation in step duration and velocity., by Daniel E. Rigobon., S.B.

Thesis: S.M., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2018., This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections., Cataloged from student-submitted PDF version of thesis., Includes bibliographical references (pages 113-119)., Despite large feedback delays, and many degrees of freedom, humans are incredibly dexterous and excel at physical interaction with complex objects. In this work we developed an upper limb crank turning experiment to study the human controller used to manage physical interaction. Subjects turned a crank with and without visual feedback, in two directions (clockwise and counterclockwise), and in three speed conditions (slow (0.075 rev/s), medium (0.5 rev/s), and fast (2 rev/s)). We made several predictions about the dependent measures including: mean speed, standard deviation of speed, coefficient of variation of speed, mean normal force, and standard deviation of normal force. We hypothesized that subjects should perform the best at slow speeds where the effect of feedback delays, inertial dynamics, and muscle noise decrease. Notably, subjects became more variable at slow speeds, and exerted significant nonzero normal force in the slow condition. At slow speeds, increased speed variability and compressive normal forces cannot be explained by biomechanics - suggesting they result from neural control. Next, the zero-force trajectory was computed. The zero-force trajectory allows for the peripheral biomechanics to be 'subtracted' to 'reveal' the underlying neural commands, expressed in terms of motion. We detected a coincidence of curvature and velocity extrema in the zero-force trajectory. Furthermore, this observation was robust to changes in impedance parameters. This finding is exciting. Even though the hand was confined to a circular path, when the peripheral biomechanics were subtracted, the same velocity curvature relationship seen in unconstrained movements was revealed. Lastly, the increased variability at slow speeds was present in the zero force trajectory. This indicates that the increased variability at slow speeds is a result of neural control, not biomechanics; this finding is consistent with previous research in unconstrained motion., National Science Foundation National Robotics Initiative Grant No. 1637824, by James Russell Hermus., S.M.

Thesis: S.M., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2018., This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections., Cataloged from student-submitted PDF version of thesis., Includes bibliographical references (pages 139-142)., This thesis documents the development of software and a control system for the InMotion2 planar robot. Software was developed to provide sensor input processing from the robot encoders and force/torque transducer, and output processing for the robot motors. A controller scheme was developed to compensate for the natural robot configuration-based inertia as well as friction and un-modeled dynamics. The inertia compensator was designed using an inertial admittance model and a nonlinear robust adaptive tracking controller based on sliding mode control. A hybrid control mode was developed in which impedance control was used to enforce a virtual constraint and inertia compensation acted along the constraint. The controller proved to be stable throughout testing and provide the desired inertia. The accuracy of the inertia compensation was within human perception limits for modest inertia references., by Jessie Thorup., S.M.

Thesis: S.M., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2018., This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections., Cataloged from student-submitted PDF version of thesis., Includes bibliographical references (pages 113-119)., Despite large feedback delays, and many degrees of freedom, humans are incredibly dexterous and excel at physical interaction with complex objects. In this work we developed an upper limb crank turning experiment to study the human controller used to manage physical interaction. Subjects turned a crank with and without visual feedback, in two directions (clockwise and counterclockwise), and in three speed conditions (slow (0.075 rev/s), medium (0.5 rev/s), and fast (2 rev/s)). We made several predictions about the dependent measures including: mean speed, standard deviation of speed, coefficient of variation of speed, mean normal force, and standard deviation of normal force. We hypothesized that subjects should perform the best at slow speeds where the effect of feedback delays, inertial dynamics, and muscle noise decrease. Notably, subjects became more variable at slow speeds, and exerted significant nonzero normal force in the slow condition. At slow speeds, increased speed variability and compressive normal forces cannot be explained by biomechanics - suggesting they result from neural control. Next, the zero-force trajectory was computed. The zero-force trajectory allows for the peripheral biomechanics to be 'subtracted' to 'reveal' the underlying neural commands, expressed in terms of motion. We detected a coincidence of curvature and velocity extrema in the zero-force trajectory. Furthermore, this observation was robust to changes in impedance parameters. This finding is exciting. Even though the hand was confined to a circular path, when the peripheral biomechanics were subtracted, the same velocity curvature relationship seen in unconstrained movements was revealed. Lastly, the increased variability at slow speeds was present in the zero force trajectory. This indicates that the increased variability at slow speeds is a result of neural control, not biomechanics; this finding is consistent with previous research in unconstrained motion., National Science Foundation National Robotics Initiative Grant No. 1637824, by James Russell Hermus., S.M.

Thesis: S.M., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2017., Cataloged from PDF version of thesis., Includes bibliographical references (pages 195-198)., This study investigates various methods by which a previously developed hand-operated actuated device is capable of measuring the mechanical impedance of both the compliant patient tissue in contact with its end effector and of the hand of the operator holding the device. The particular device being investigated is an actuated ultrasound probe originally designed to regulate the amount of force exerted by a sonographer to their patient during an ultrasound scan. We expect that quantifying the effective mechanical impedance of the operator hand will lead to improvements in the design and control of hand-operated devices. Improvements are being considered not only to improve the quality and reliability of the ultrasound scan, but to alleviate the chronic muscular and joint stress endured by sonographers over the course of their careers. An additional motivation behind this study is to augment the capabilities of physicians to diagnose medical conditions, such as breast cancer and liver cirrhosis, on the basis of tissue impedance without the need for an additional mechanism. Several methods were developed for approximating mechanical impedance using the actuated ultrasound probe, based on the sensors available and models of the device dynamics. Due to sensor limitations, impedance measurement could only be effectively implemented for a single interface, instead of both concurrently. These methods involved immobilizing one end of the device. Experiments were conducted on artificial tissues in order to confirm that the methods developed were valid and reliable for measuring mechanical impedance of the body in contact with either the sonographer or patient end of the device., by David Mercado., S.M.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2016., Cataloged from PDF version of thesis., Includes bibliographical references (pages 161-175)., This thesis investigates the role of foot-ground dynamics in robot legged locomotion. Empirical observation with robot hardware is critical for such investigation due to the complexity of impact dynamics and the infinite variability of real terrain. To this end, we introduce the MIT Super Mini Cheetah, a small, inexpensive quadrupedal robot that can run and walk over a variety of ground types (gravel, tile, carpet, grass, etc.). Experiments with the Super Mini Cheetah demonstrate new terrain-adaptation capabilities for locomotion control over surfaces with varying mechanical impedance. The terrain adaptation methods are shown empirically to increase gait stability and to reduce impact loading on the legs. To the best of our knowledge, the work demonstrates the first example of a running controller that measures changes in stiffness of the ground and modifies its controller within a single stride. This enables the Super Mini Cheetah to run smoothly between hard and soft surfaces. Algorithms to quickly measure ground properties are presented. These measurements are made by observing interaction between robot legs and the ground. We show that the impact event between the foot and the ground is particularly information-rich which enables the Super Mini Cheetah to estimate ground stiffness within 50 milliseconds of beginning a step. These results contribute new knowledge about the hardware and control requirements to provide legged robots with a useful sense of touch, which will greatly enhance the types of terrain that legged robots can safely and gracefully roam., by Will Bosworth., Ph. D.

Thesis: S.M., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2017., Cataloged from PDF version of thesis., Includes bibliographical references (pages 80-82)., Agility is defined as the ability to quickly change speed or direction. Planned agility refers to the physical act of changing direction and reactive agility addresses the additional cognitive responses needed to react quickly to an external cue. This work specifically considers reactive agility. Agility performance is often evaluated using time-based metrics, which provide little information about which factors aid or limit success. Two studies were completed to identify key factors contributing to agility performance. The objective of the first study was to determine how novices and experts working in athletic, clinical, and military environments qualitatively and quantitatively evaluate agility performance. Thirty-three participants completed a survey which involved scoring 16 athletes on a 7 point Likert scale of not agile to agile. The spread of the scores indicated that even within groups, participants had different opinions about which aspects of technique contributed to high performance. Participant responses were used to link several terms to agility technique. The objective of the second study was to apply these terms to the development of objective biomechanical metrics. An array of body-worn inertial sensors was used to calculate metrics that were sensitive to performance speed. Five metrics were defined (normalized number of foot contacts, stride length variance, arm swing variance, mean normalized stride frequency, and number of body rotations). Eighteen participants donned 13 sensors to complete a reactive agility task, which involved navigating a set of cones in response to a vocal cue. Participants were grouped into fast, medium, and slow performance based on their completion time. Participants in the fast group had the smallest number of foot contacts after normalizing by height, highest stride length variance, highest forearm angular velocity variance, and highest stride frequency after normalizing by height.These metric values translate to an eff, by Chika U. Eke., S.M.

Thesis: S.M., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2017., Cataloged from PDF version of thesis., Includes bibliographical references (pages 195-198)., This study investigates various methods by which a previously developed hand-operated actuated device is capable of measuring the mechanical impedance of both the compliant patient tissue in contact with its end effector and of the hand of the operator holding the device. The particular device being investigated is an actuated ultrasound probe originally designed to regulate the amount of force exerted by a sonographer to their patient during an ultrasound scan. We expect that quantifying the effective mechanical impedance of the operator hand will lead to improvements in the design and control of hand-operated devices. Improvements are being considered not only to improve the quality and reliability of the ultrasound scan, but to alleviate the chronic muscular and joint stress endured by sonographers over the course of their careers. An additional motivation behind this study is to augment the capabilities of physicians to diagnose medical conditions, such as breast cancer and liver cirrhosis, on the basis of tissue impedance without the need for an additional mechanism. Several methods were developed for approximating mechanical impedance using the actuated ultrasound probe, based on the sensors available and models of the device dynamics. Due to sensor limitations, impedance measurement could only be effectively implemented for a single interface, instead of both concurrently. These methods involved immobilizing one end of the device. Experiments were conducted on artificial tissues in order to confirm that the methods developed were valid and reliable for measuring mechanical impedance of the body in contact with either the sonographer or patient end of the device., by David Mercado., S.M.

Thesis: S.M., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2016., Cataloged from PDF version of thesis., Includes bibliographical references (pages 99-102)., This thesis explores the use of impedance control on an anthropomorphic robot for operations in extreme, poorly mapped environments. First, a dynamic model was developed for a Baxter Research Robot. This model improved on standard dynamic models for similar robots by including the dynamics of the actuators in the system. Specifically, it was demonstrated that when the effective inertia of the actuators is neglected, the system will transmit 1.6 times more force to the environment than the model predicts. A force based Cartesian impedance controller was then implemented on Baxter, and numerous ways to modulate the endpoint impedance, including feedback and geometric configuration, were discussed and compared. Finally, a series of scaled down tasks similar to ones which are required in the decommissioning of offshore oil fields were then completed on Baxter using the Cartesian impedance controller. Overall, it was demonstrated that by using this more advanced control scheme, Baxter was (1) capable of satisfactorily completing the scaled down tasks, (2) more robust against errors in the map of the environment than with traditional controllers, and (3) capable of improving the map of its environment while completing the task., by Lucille Aileen Hosford., S.M.

Thesis: S.B., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2016., Cataloged from PDF version of thesis., Includes bibliographical references (page 42)., This thesis aims to investigate human motor control strategies. Curved constraints offer a unique opportunity to exploit forces of contact. A circular crank experiment using the MIT MANUS robot was designed in order to test how well subjects can follow a set of simple instructions to rotate the crank at various constant speeds. 10 subjects volunteered to participate in this experiment. Velocity, force, and EMG data were collected during four tasks: turning the crank at the subject's preferred or comfortable speed, turning the crank at a constant preferred speed, turning the crank at a constant preferred speed with a visual feedback display, and rotating the crank at three instructed speeds (slow, medium, and fast) with visual feedback. The coefficient of variation (CV) of the velocity for each trial was computed as a measure of performance. Statistical analysis showed that speed significantly affected CV but the direction of turning the crank, clockwise or counterclockwise, did not. The observation that CV increased as speed decreased, despite visual feedback, confirms previous studies showing that human motor control is more imprecise at slower speeds., by Brian Wilcox., S.B.

Thesis: S.M., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2016., This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections., Cataloged from student-submitted PDF version of thesis., Includes bibliographical references (pages 94-104)., Rehabilitation of human motor function is an issue of utmost significance affecting millions of Americans. Robot-aided therapy emerged as a promising method to meet the increasing demand for effective rehabilitation services. While robot-aided therapy for upper-extremity provides clinically-proven, efficient rehabilitation, human-interactive robots for lower-extremity therapy have been substantially less successful. Given the labor-intensive nature of conventional, human-administered walking therapy, effective robot-aided assistance is urgently needed. The use of robots and treadmills that may inadvertently suppress the expression of the natural oscillatory dynamics of walking is addressed in this thesis as a possible explanation for the ineffectiveness of robotic walking therapy. To further investigate the natural oscillatory dynamics of walking, the existence and provenance (spinal or central) of a neuro-mechanical oscillator underlying human locomotion was assessed. This oscillator was studied via gait entrainment to periodic mechanical perturbations at the ankle in both treadmill and overground environments. Experiments with unimpaired human subjects provided direct behavioral evidence of the non-negligible contribution to human walking made by a limit-cycle oscillator in the spinal neuro-mechanical periphery. Entrainment was always accompanied by phase-locking so that plantar-flexion perturbations assisted propulsion during ankle 'push-off' while dorsi-flexion perturbations assisted toe-clearance during 'initial swing'. The observed behavior seemed to require a neural adaptation that could not easily be ascribed to biomechanics, suggesting a hierarchical organization between the supra-spinal nervous system and the spinal neuro-mechanical periphery: episodic supervisory control., by Julieth Ochoa., S.M.

Thesis: S.B., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2016., Cataloged from PDF version of thesis., Includes bibliographical references (page 45)., Robotic control of complex objects is inferior to that of humans despite superior communication, sensors and actuators. Therefore, studying human control of complex dynamic objects, in particular a bullwhip, should reveal how humans may achieve superior dexterity. An expert whip-cracker performing two distinct targeting tasks, discrete and rhythmic, was observed using the Qualisys 3D motion capture software. The objective was to investigate the kinematics of the whip, the kinematics of the subject's arm while controlling the whip and the differences between discrete and rhythmic tasks. The subject was able to expertly perform both targeting tasks with excellent targeting accuracy. The study confirmed the existence of a wave propagating down the whip as stated in prior work. Furthermore, a distinct difference between discrete and rhythmic tasks was observed in the reproducibility of position profiles, reproducibility of phase profiles, and the waveforms of elbow and wrist angles. Finally, the whip trajectory was substantially confined to a plane. In contrast with claims made in prior work, this plane was found to be distinct from the parasagittal plane and slanted with respect to it., by Camille Henrot., S.B.

Thesis: S.B., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2016., Cataloged from PDF version of thesis., Includes bibliographical references (page 42)., This thesis aims to investigate human motor control strategies. Curved constraints offer a unique opportunity to exploit forces of contact. A circular crank experiment using the MIT MANUS robot was designed in order to test how well subjects can follow a set of simple instructions to rotate the crank at various constant speeds. 10 subjects volunteered to participate in this experiment. Velocity, force, and EMG data were collected during four tasks: turning the crank at the subject's preferred or comfortable speed, turning the crank at a constant preferred speed, turning the crank at a constant preferred speed with a visual feedback display, and rotating the crank at three instructed speeds (slow, medium, and fast) with visual feedback. The coefficient of variation (CV) of the velocity for each trial was computed as a measure of performance. Statistical analysis showed that speed significantly affected CV but the direction of turning the crank, clockwise or counterclockwise, did not. The observation that CV increased as speed decreased, despite visual feedback, confirms previous studies showing that human motor control is more imprecise at slower speeds., by Brian Wilcox., S.B.

Thesis: S.M., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2016., This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections., Cataloged from student-submitted PDF version of thesis., Includes bibliographical references (pages 94-104)., Rehabilitation of human motor function is an issue of utmost significance affecting millions of Americans. Robot-aided therapy emerged as a promising method to meet the increasing demand for effective rehabilitation services. While robot-aided therapy for upper-extremity provides clinically-proven, efficient rehabilitation, human-interactive robots for lower-extremity therapy have been substantially less successful. Given the labor-intensive nature of conventional, human-administered walking therapy, effective robot-aided assistance is urgently needed. The use of robots and treadmills that may inadvertently suppress the expression of the natural oscillatory dynamics of walking is addressed in this thesis as a possible explanation for the ineffectiveness of robotic walking therapy. To further investigate the natural oscillatory dynamics of walking, the existence and provenance (spinal or central) of a neuro-mechanical oscillator underlying human locomotion was assessed. This oscillator was studied via gait entrainment to periodic mechanical perturbations at the ankle in both treadmill and overground environments. Experiments with unimpaired human subjects provided direct behavioral evidence of the non-negligible contribution to human walking made by a limit-cycle oscillator in the spinal neuro-mechanical periphery. Entrainment was always accompanied by phase-locking so that plantar-flexion perturbations assisted propulsion during ankle 'push-off' while dorsi-flexion perturbations assisted toe-clearance during 'initial swing'. The observed behavior seemed to require a neural adaptation that could not easily be ascribed to biomechanics, suggesting a hierarchical organization between the supra-spinal nervous system and the spinal neuro-mechanical periphery: episodic supervisory control., by Julieth Ochoa., S.M.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 1985., MICROFICHE COPY AVAILABLE IN ARCHIVES AND ENGINEERING, Bibliography: leaf 111., by David L. Dean, Jr., M.S.

Thesis: S.B., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2015., This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections., Cataloged from student-submitted PDF version of thesis., Includes bibliographical references (page 64)., The InMotion² robot is a clinical version of the MIT-Manus robot, which was developed in the 1990s to help with upper limb rehabilitation of stroke survivors. In 2015, experiments were planned to use the InMotion² robot for studies of human force and motion control. During preparation for the experiments, however, a malfunction was discovered in the robot. A series of systematic tests were carried out to determine what part of the robot was causing the malfunction. It was determined that the magnets on one of the two motors were slipping on the rotor shaft. This slippage caused the malfunctioning motor's torque output to range from 10% to 13% of the other motor's output given the same input signal. The malfunctioning motor was repaired, and the robot was reassembled. Tests were carried out to verify the performance of the robot, and the torque output of the malfunctioning motor was measured to range from 120% to 130% of the other motor's torque output, showing that the malfunction had been fixed., by Margaret Mary Coad., S.B.

Thesis: S.B., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2015., This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections., Cataloged from student-submitted PDF version of thesis., Includes bibliographical references (page 64)., The InMotion² robot is a clinical version of the MIT-Manus robot, which was developed in the 1990s to help with upper limb rehabilitation of stroke survivors. In 2015, experiments were planned to use the InMotion² robot for studies of human force and motion control. During preparation for the experiments, however, a malfunction was discovered in the robot. A series of systematic tests were carried out to determine what part of the robot was causing the malfunction. It was determined that the magnets on one of the two motors were slipping on the rotor shaft. This slippage caused the malfunctioning motor's torque output to range from 10% to 13% of the other motor's output given the same input signal. The malfunctioning motor was repaired, and the robot was reassembled. Tests were carried out to verify the performance of the robot, and the torque output of the malfunctioning motor was measured to range from 120% to 130% of the other motor's torque output, showing that the malfunction had been fixed., by Margaret Mary Coad., S.B.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2000., Includes bibliographical references (p. 121-124)., by Max Berniker., S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2002., Includes bibliographical references (p. 143-145)., by Philip H. Tang., S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2001., Includes bibliographical references (p. 179-181)., by Stephen Paul Buerger., S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2001., Includes bibliographical references (p. 179-181)., by Stephen Paul Buerger., S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 1999., Includes bibliographical references (leaves 125-126)., by Craig Foster., S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 1999., Includes bibliographical references (leaves 125-126)., by Craig Foster., S.M.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2013., Cataloged from PDF version of thesis., Includes bibliographical references (p. 222-230)., Ankle mechanical impedance, which is a dynamic relationship between angular displacement and the corresponding torque at the ankle joint, plays a key role in natural interaction of the lower-extremity with the environment. The human ankle is a biomechanically complex joint consisting of three bones with non-intersecting anatomical axes, and its motions under normal motor control and function are predominantly in multiple degrees-of-freedom (DOF). This thesis provides a quantitative characterization of multivariable ankle mechanical impedance of young healthy subjects in two DOF, both in the sagittal and the frontal planes. Multi-variable studies provide several important characteristics of the human ankle, unavailable from single DOF studies, which have mostly been in the sagittal plane. Three characterization methods were developed to study ankle mechanical impedance in different conditions: 1) steady-state static, 2) steady-state dynamic, and 3) transient dynamic. First, steady-state static ankle mechanical impedance, which is a non-linear torque and angle relationship at the ankle, was characterized in two coupled DOFs over the normal range of motion. Robust vector field approximation methods based on thin-plate spline smoothing with generalized cross validation showed that static ankle impedance is highly direction dependent, being weak in the inversion-eversion direction. Activating a single muscle or co-contracting antagonistic muscles significantly increased static ankle impedance in all directions but more in the dorsiflexion-plantarflexion direction than the inversion-eversion. Static ankle behavior in both relaxed and active muscles was close to that of a passive elastic system. Second, steady-state dynamic ankle mechanical impedance was characterized based on linear time-invariant multi-input multi-output stochastic system identification methods. A highly linear relationship between muscle activation and ankle impedance was identified in all movement dire, by Hyunglae Lee., Ph.D.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2012., Cataloged from PDF version of thesis., Includes bibliographical references (p. 51)., The biomechanics of lower limb locomotion is a yet unknown mixture of neurological control and physical parameters. The current study explored attaching a rehabilitative anklebot to subjects walking on a treadmill and observed duration, kinematic, and electromyography data to determine the biomechanical response to the asymmetric loading. The present report identified various gait cycle parameters that changed as a response to the asymmetric loading. Notably, significant differences in the stride time of the legs occurred under loading, while contralateral stride times also adjusted to remain equal to those of the loaded legs. Symmetry index analysis led to the conclusion that, while the asymmetric loading of the lower limbs had some effects on temporal gait parameters, the body adjusted to minimize any temporal asymmetry. However, goniometer data demonstrated kinematic changes in response to loading as knee flexion peaked earlier in the gait cycle., by Wesley D. McDougal., S.B.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2011., Cataloged from PDF version of thesis., Includes bibliographical references (p. 51)., This project used an exoskeletal robot to increase and decrease the stiffness of the ankle joint during treadmill walking to measure the effect of ankle impedance on lower limb joint kinematics. By quantifying the effect of ankle impedance on the knee joint we sought to better understand coordination and control of the ankle and knee. Using linear regression to determine the relationship between the maximum knee flexion during stance and the imposed stiffness on the ankle, we found a measurable positive correlation in 4 out of 5 test subjects at a 95% confidence level. The knee responded to modifications in ankle stiffness as expected from a simple mechanical model. Remarkably, the response was small and variable enough to suggest the body compensates to preserve normal kinematic profiles., by Bonnie Lucille Blackburn., S.B.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2012., Cataloged from PDF version of thesis., Includes bibliographical references (p. 51)., The biomechanics of lower limb locomotion is a yet unknown mixture of neurological control and physical parameters. The current study explored attaching a rehabilitative anklebot to subjects walking on a treadmill and observed duration, kinematic, and electromyography data to determine the biomechanical response to the asymmetric loading. The present report identified various gait cycle parameters that changed as a response to the asymmetric loading. Notably, significant differences in the stride time of the legs occurred under loading, while contralateral stride times also adjusted to remain equal to those of the loaded legs. Symmetry index analysis led to the conclusion that, while the asymmetric loading of the lower limbs had some effects on temporal gait parameters, the body adjusted to minimize any temporal asymmetry. However, goniometer data demonstrated kinematic changes in response to loading as knee flexion peaked earlier in the gait cycle., by Wesley D. McDougal., S.B.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2011., Cataloged from PDF version of thesis., Includes bibliographical references (p. 51)., This project used an exoskeletal robot to increase and decrease the stiffness of the ankle joint during treadmill walking to measure the effect of ankle impedance on lower limb joint kinematics. By quantifying the effect of ankle impedance on the knee joint we sought to better understand coordination and control of the ankle and knee. Using linear regression to determine the relationship between the maximum knee flexion during stance and the imposed stiffness on the ankle, we found a measurable positive correlation in 4 out of 5 test subjects at a 95% confidence level. The knee responded to modifications in ankle stiffness as expected from a simple mechanical model. Remarkably, the response was small and variable enough to suggest the body compensates to preserve normal kinematic profiles., by Bonnie Lucille Blackburn., S.B.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2012., Cataloged from PDF version of thesis., Includes bibliographical references (p. 33)., Computing environments such as Matlab that are conventionally used to simulate dynamics of rigid body systems can be used to model interactions between the system and its environment. However, creating these simulations using Matlab or an equivalent is difficult and there is a need for a more convenient simulation environment for such problems. Two alternative programs, PyODE and OpenSim, were explored to evaluate their ability to fill this need. Models and simulations of the human ankle were created in PyODE. This program is useful for creating simple models where the programmer desires a high level of control over model parameters. Simulations of the ankle kicking a ball and taking a step were created to examine the effect of joint stiffness on these motions and help determine the usefulness of ODE as a simulation tool. Pre-existing models were analyzed in OpenSim. OpenSim is specifically designed for analyzing biomechanical systems. It allows for more complex models to be created but the user has more limited control over the model parameters., by Rebecca Vasquez., S.B.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2012., Cataloged from PDF version of thesis., Includes bibliographical references (p. 214-226)., Growing interest in robotic treatment of patients with neurological injury motivates the development of therapeutic robots for basic research into recovery. Though humans are the ultimate beneficiaries, basic research frequently involves rodent models of neurological injury, which motivates robotic devices that can interact with rats or mice. Currently, available apparatus for locomotion studies of rodents is built upon treadmills, which simplify the design and implementation but also restrict the scope of possible experiments. This is largely due to the treadmill's single-dimensional movement and the lack of accommodation for natural or voluntary movement of the animal. In order to open up new possibilities for locomotion studies in rodents, this work introduces newly developed apparatus for locomotion research in rodents. The key concept is to allow maximal freedom of voluntary movement of the animal while providing forceful interaction when necessary. Advantages and challenges of the proposed machine over other existing designs are discussed. Design and implementation issues are presented and discussed, emphasizing their impact on free, voluntary, movement of the animal. A live-animal experiment was conducted to verify the design principles. Unconstrained natural movement of the animal was compared with movement with the overground robot attached. The compact, overground design and backdrivable implementation of this robot allow novel experiments that involve open-space, free (or interactive) locomotion of the animal., by Yun Seong Song., Ph.D.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2012., Cataloged from PDF version of thesis., Includes bibliographical references (p. 33)., Computing environments such as Matlab that are conventionally used to simulate dynamics of rigid body systems can be used to model interactions between the system and its environment. However, creating these simulations using Matlab or an equivalent is difficult and there is a need for a more convenient simulation environment for such problems. Two alternative programs, PyODE and OpenSim, were explored to evaluate their ability to fill this need. Models and simulations of the human ankle were created in PyODE. This program is useful for creating simple models where the programmer desires a high level of control over model parameters. Simulations of the ankle kicking a ball and taking a step were created to examine the effect of joint stiffness on these motions and help determine the usefulness of ODE as a simulation tool. Pre-existing models were analyzed in OpenSim. OpenSim is specifically designed for analyzing biomechanical systems. It allows for more complex models to be created but the user has more limited control over the model parameters., by Rebecca Vasquez., S.B.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2011., Cataloged from PDF version of thesis., Includes bibliographical references (p. 149-156)., Rehabilitation of human motor function is an issue of the utmost significance, and the demand for the effective rehabilitation service is even growing with the graying of the population. Robotic technology has provided promising ways to assist recovery of the motor function of upper extremities. In contrast, current robotic therapy for lower extremities has shown inferior efficacy. In this thesis, the source of the limited efficacy of current robotic walking therapy is addressed. Essential mechanical components for robustly stable walking are identified as energy dissipation and proper compensation. Based on these essential components, design criteria of effective robotic walking therapy are suggested as foot-ground interaction and ankle actuation. A novel strategy of robot aided walking therapy reflecting the design criteria is proposed; dynamically entraining human gait with periodic ankle torque from a robot. Experiments with normal subjects and neurologically impaired subjects support the feasibility of the proposed rehabilitation strategy. The gait period of subjects entrain to the periodic mechanical perturbation with a measurable basin of entrainment, and the entrainment always accompanies phase-locking so that the mechanical perturbation assists propulsion. These observations are affected neither by auditory feedback nor by a distractor task for normal subjects, and consistently observed in impaired subjects. A highly simplified one degree of freedom walking model without supra-spinal control or an intrinsic self-sustaining neural oscillator (a rhythmic pattern generator) encapsulated the essence of these observations. This suggests that several prominent limit-cycle features of human walking may stem from peripheral mechanics mediated by simple afferent feedback without significant involvement of supra-spinal control or central pattern generator. The competence of the highly simplified model supports that the proposed entrainment therapy may be effective fo, by Jooeun Ahn., Ph.D.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2010., Cataloged from PDF version of thesis., Includes bibliographical references (p. 59)., In this thesis, I conducted an in-vivo study providing measurements of human static ankle mechanical impedance. Accurate measurements of ankle impedance when muscles were voluntarily activated were obtained using a therapeutic robot, Anklebot, and an electromyographic recording system. Important features of ankle impedance, and their variation with muscle activity, are discussed, including magnitude, symmetry and directions of minimum and maximum impedance. Voluntary muscle activation has a significant impact on ankle impedance, increasing it by up to a factor of three in our experiments. Furthermore, significant asymmetries and deviations from a linear two-spring model are present in many subjects, indicating that ankle impedance has a complex and individually idiosyncratic structure. I propose the use of Fourier series as a general representation, providing both insight and a precise quantitative characterization of human static ankle impedance., by Patrick Ho., S.M.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2011., Cataloged from PDF version of thesis., Includes bibliographical references (p. 149-156)., Rehabilitation of human motor function is an issue of the utmost significance, and the demand for the effective rehabilitation service is even growing with the graying of the population. Robotic technology has provided promising ways to assist recovery of the motor function of upper extremities. In contrast, current robotic therapy for lower extremities has shown inferior efficacy. In this thesis, the source of the limited efficacy of current robotic walking therapy is addressed. Essential mechanical components for robustly stable walking are identified as energy dissipation and proper compensation. Based on these essential components, design criteria of effective robotic walking therapy are suggested as foot-ground interaction and ankle actuation. A novel strategy of robot aided walking therapy reflecting the design criteria is proposed; dynamically entraining human gait with periodic ankle torque from a robot. Experiments with normal subjects and neurologically impaired subjects support the feasibility of the proposed rehabilitation strategy. The gait period of subjects entrain to the periodic mechanical perturbation with a measurable basin of entrainment, and the entrainment always accompanies phase-locking so that the mechanical perturbation assists propulsion. These observations are affected neither by auditory feedback nor by a distractor task for normal subjects, and consistently observed in impaired subjects. A highly simplified one degree of freedom walking model without supra-spinal control or an intrinsic self-sustaining neural oscillator (a rhythmic pattern generator) encapsulated the essence of these observations. This suggests that several prominent limit-cycle features of human walking may stem from peripheral mechanics mediated by simple afferent feedback without significant involvement of supra-spinal control or central pattern generator. The competence of the highly simplified model supports that the proposed entrainment therapy may be effective fo, by Jooeun Ahn., Ph.D.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2010., Cataloged from PDF version of thesis., In this thesis, a series viscous actuator, composed of a viscous rotary damper connected to the output of a geared electric motor, is proposed as a novel solution to the high actuator impedances encountered when using such a motor. High impedances can be undesirable when interacting with the environment or animal subjects as unexpected forces may damage already-compromised joints. A state space model for the series viscous actuator was derived and verified by measurements upon a physical system. Physical design considerations for implementing a series viscous actuator, such as the appropriate gearing ratio and damper selection, are discussed and supported using the derived model., by Nicholas Eric Wiltsie., S.B.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2010., Cataloged from PDF version of thesis., Includes bibliographical references (p. 23)., As part of an ongoing effort to better understand and treat locomotor disorders, an over-ground therapeutic robot prototype to study recovery of locomotion after spinal cord injury in rodents is under development. One key element of the therapeutic robot is a system to support the partial body weight of a freely-moving rodent. This paper discusses the design requirements, fabrication, modeling, calibration and preliminary analysis of a highly back-drivable bodyweight support system prototype. In addition, a closed loop feedback control system was designed, simulated, constructed and tested. Hardware limitations were identified, and alternative control techniques were explored., by Michaëlle Ntala Mayalu., S.B.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2010., Cataloged from PDF version of thesis., In this thesis, a series viscous actuator, composed of a viscous rotary damper connected to the output of a geared electric motor, is proposed as a novel solution to the high actuator impedances encountered when using such a motor. High impedances can be undesirable when interacting with the environment or animal subjects as unexpected forces may damage already-compromised joints. A state space model for the series viscous actuator was derived and verified by measurements upon a physical system. Physical design considerations for implementing a series viscous actuator, such as the appropriate gearing ratio and damper selection, are discussed and supported using the derived model., by Nicholas Eric Wiltsie., S.B.

Thesis (S.B.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2010., Cataloged from PDF version of thesis., In this thesis, a series viscous actuator, composed of a viscous rotary damper connected to the output of a geared electric motor, is proposed as a novel solution to the high actuator impedances encountered when using such a motor. High impedances can be undesirable when interacting with the environment or animal subjects as unexpected forces may damage already-compromised joints. A state space model for the series viscous actuator was derived and verified by measurements upon a physical system. Physical design considerations for implementing a series viscous actuator, such as the appropriate gearing ratio and damper selection, are discussed and supported using the derived model., by Nicholas Eric Wiltsie., S.B.

Thesis (Ph. D.)--Harvard-MIT Division of Health Sciences and Technology, 2008., This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections., Includes bibliographical references (p. 173-178)., Over the past three decades, much research in motor neuroscience has focused on understanding how humans make coordinated reaching movements, yielding valuable insight into the planning and control of reaching movements, and establishing a foundation for robot-assisted rehabilitation. The goal of this doctoral research was to provide a quantitative characterization of humans' wrist rotations, paving the way for intelligent robot-assisted wrist rehabilitation. More specifically, we have characterized the kinematics, dynamics, and adaptation of wrist rotations, and discussed implications for planning and control. Kinematics: It is well known that humans make relatively straight reaching movements, suggesting that reaching movements are primarily under kinematic control of hand position. We used a motion capture system to test if wrist rotations are also under kinematic control. We found that wrist rotations exhibit a pattern with significantly more path curvature and variability than reaching movements (p = 0.001). While the increased path curvature could indicate that wrist rotations are not under kinematic control, this work provides evidence that the curvature is instead due to imperfect peripheral execution.Dynamics: In order to determine the exact cause of path curvature, an anatomically-accurate, mathematical model of the wrist was developed, including recent measurements of passive wrist stiffness. Combining experimentally-measured kinematics from human subjects with the wrist model revealed that moderately-sized wrist rotations can be approximated by a very simple model with virtually no loss in accuracy.Interaction torques, for which the nervous system compensates in reaching movements, are present but negligible in wrist rotations., (cont) Rather, wrist rotation dynamics are dominated by stiffness, which was shown to be the likely cause of path curvature.Adaptation: When perturbed during reaching movements, humans adapt by straightening their paths, confirming that kinematics play a prominent role in planning reaching movements. We found that subjects consistently adapted to a conservative,velocity-dependent force field. Interestingly, this adaptation was more difficult to detect than in perturbation studies involving reaching movements. Taken together, these results suggest that wrist rotations are also primarily under kinematic control (albeit imperfect)., by Steven K. Charles., Ph.D.

Thesis (Ph. D.)--Harvard-MIT Division of Health Sciences and Technology, 2008., This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections., Includes bibliographical references (p. 173-178)., Over the past three decades, much research in motor neuroscience has focused on understanding how humans make coordinated reaching movements, yielding valuable insight into the planning and control of reaching movements, and establishing a foundation for robot-assisted rehabilitation. The goal of this doctoral research was to provide a quantitative characterization of humans' wrist rotations, paving the way for intelligent robot-assisted wrist rehabilitation. More specifically, we have characterized the kinematics, dynamics, and adaptation of wrist rotations, and discussed implications for planning and control. Kinematics: It is well known that humans make relatively straight reaching movements, suggesting that reaching movements are primarily under kinematic control of hand position. We used a motion capture system to test if wrist rotations are also under kinematic control. We found that wrist rotations exhibit a pattern with significantly more path curvature and variability than reaching movements (p = 0.001). While the increased path curvature could indicate that wrist rotations are not under kinematic control, this work provides evidence that the curvature is instead due to imperfect peripheral execution.Dynamics: In order to determine the exact cause of path curvature, an anatomically-accurate, mathematical model of the wrist was developed, including recent measurements of passive wrist stiffness. Combining experimentally-measured kinematics from human subjects with the wrist model revealed that moderately-sized wrist rotations can be approximated by a very simple model with virtually no loss in accuracy.Interaction torques, for which the nervous system compensates in reaching movements, are present but negligible in wrist rotations., (cont) Rather, wrist rotation dynamics are dominated by stiffness, which was shown to be the likely cause of path curvature.Adaptation: When perturbed during reaching movements, humans adapt by straightening their paths, confirming that kinematics play a prominent role in planning reaching movements. We found that subjects consistently adapted to a conservative,velocity-dependent force field. Interestingly, this adaptation was more difficult to detect than in perturbation studies involving reaching movements. Taken together, these results suggest that wrist rotations are also primarily under kinematic control (albeit imperfect)., by Steven K. Charles., Ph.D.

Thesis (Ph. D.)--Harvard-MIT Division of Health Sciences and Technology, 2008., This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections., Includes bibliographical references (p. 173-178)., Over the past three decades, much research in motor neuroscience has focused on understanding how humans make coordinated reaching movements, yielding valuable insight into the planning and control of reaching movements, and establishing a foundation for robot-assisted rehabilitation. The goal of this doctoral research was to provide a quantitative characterization of humans' wrist rotations, paving the way for intelligent robot-assisted wrist rehabilitation. More specifically, we have characterized the kinematics, dynamics, and adaptation of wrist rotations, and discussed implications for planning and control. Kinematics: It is well known that humans make relatively straight reaching movements, suggesting that reaching movements are primarily under kinematic control of hand position. We used a motion capture system to test if wrist rotations are also under kinematic control. We found that wrist rotations exhibit a pattern with significantly more path curvature and variability than reaching movements (p = 0.001). While the increased path curvature could indicate that wrist rotations are not under kinematic control, this work provides evidence that the curvature is instead due to imperfect peripheral execution.Dynamics: In order to determine the exact cause of path curvature, an anatomically-accurate, mathematical model of the wrist was developed, including recent measurements of passive wrist stiffness. Combining experimentally-measured kinematics from human subjects with the wrist model revealed that moderately-sized wrist rotations can be approximated by a very simple model with virtually no loss in accuracy.Interaction torques, for which the nervous system compensates in reaching movements, are present but negligible in wrist rotations., (cont) Rather, wrist rotation dynamics are dominated by stiffness, which was shown to be the likely cause of path curvature.Adaptation: When perturbed during reaching movements, humans adapt by straightening their paths, confirming that kinematics play a prominent role in planning reaching movements. We found that subjects consistently adapted to a conservative,velocity-dependent force field. Interestingly, this adaptation was more difficult to detect than in perturbation studies involving reaching movements. Taken together, these results suggest that wrist rotations are also primarily under kinematic control (albeit imperfect)., by Steven K. Charles., Ph.D.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2006., Includes bibliographical references (p. 181-182)., Experimental data show that ankle torque is the most important actuator in normal human locomotion. I investigate the dynamics of simple models actuated by ankles alone. To assess the contribution of ankle actuation to locomotion, I first analyze the dynamics of some passive walkers without any joint torque. These passive walkers include a rimless wheel model and springy-legged models with and without a double stance phase. I analyze the stability of the period-one gait of each passive walker to compare it with the stability of the period-one gait of an ankle actuated model. Subsequently, I investigate whether balancing of a double inverted pendulum model whose shape and mass distribution are similar to a human can be achieved by control of ankle torque in a frontal plane. I study the dynamics of the model and design a controller that makes the model balance with biologically realistic ankle torque and a reasonable foot-floor friction coefficient. I conclude that an ankle-actuated model can make a stable period-one gait in a sagittal plane. Also, I deduce that the ankle torque control in a frontal plane can stabilize a double inverted pendulum model whose shape and mechanical properties are similar to those of humans., by Jooeun Ahn., S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2006., Includes bibliographical references (leaves 187-189)., The aim of this thesis was to further characterize the effectiveness of field responsive fluids (FRFs) in geometries pertinent to the soldier and to examine the effects of specific geometric and kinematic parameters, including patterned surface geometry, electrode gap distance, and normal force on the performance of homogeneous ERF composites. Field responsive fluid composites designed for variable impedance energy absorption incorporated electrorheological fluid (ERF) and shear-thickening fluid (STF) in novel geometries to absorb compressive and tensile/shear forces. ER and ST fluids change their apparent viscosity in the presence of elevated electric and shear fields, respectively, and the magnitude of this effect can be adjusted using the magnitude of the input field, allowing variable impedance operation. Several test fixtures were developed to test these novel FRF composites. A compression apparatus was designed and constructed to test STF-filled foam over a range of strain rates not previously examined in the literature. Silicon-based microchannel devices with etched features on the order of 100 pm and etch depths of 7-90 pm were fabricated to test homogeneous ER fluids in small electrode gaps., (cont.) Tests using these silicon devices allowed creation of 5 kV/mm (5 V/pm) electric fields across electrode gaps as small as 20 pm, with increases of measured shear force as high as 350% from no electric field to full 5 kV/mm operation. Production of these devices in bulk using established silicon processing techniques was demonstrated, and factors affecting the manufacture of these devices were investigated., by Ryan A. Griffin., S.M.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 1988., Bibliography: leaves 201-206., by Wyatt S. Newman., Ph.D.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2006., Includes bibliographical references (p. 181-182)., Experimental data show that ankle torque is the most important actuator in normal human locomotion. I investigate the dynamics of simple models actuated by ankles alone. To assess the contribution of ankle actuation to locomotion, I first analyze the dynamics of some passive walkers without any joint torque. These passive walkers include a rimless wheel model and springy-legged models with and without a double stance phase. I analyze the stability of the period-one gait of each passive walker to compare it with the stability of the period-one gait of an ankle actuated model. Subsequently, I investigate whether balancing of a double inverted pendulum model whose shape and mass distribution are similar to a human can be achieved by control of ankle torque in a frontal plane. I study the dynamics of the model and design a controller that makes the model balance with biologically realistic ankle torque and a reasonable foot-floor friction coefficient. I conclude that an ankle-actuated model can make a stable period-one gait in a sagittal plane. Also, I deduce that the ankle torque control in a frontal plane can stabilize a double inverted pendulum model whose shape and mechanical properties are similar to those of humans., by Jooeun Ahn., S.M.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 1988., Bibliography: leaves 201-206., by Wyatt S. Newman., Ph.D.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2005., Includes bibliographical references (p. 335-343)., Each year, about 700,000 people in the United States have a stroke, making it a leading cause of serious, long-term disability. Modalities of therapy often assume the processes underlying motor recovery and motor learning are similar because both exhibit activity- dependent neural plasticity. However, the impact of other factors unique to recovery such as re-acquisition of muscle strength and resolution of abnormal muscle tone confounds the validity of this assumption. By implementing an adaptive impedance controller that collapses from a "virtual slot" between two targets to a "virtual spring" at the desired target, a new performance-based progressive therapy (PBPT) algorithm was developed to test whether recovery would be enhanced by incorporating learning strategies like repetition, goal specification, and positive reinforcement. A study of chronic stroke patients (8 to 95 months post-stroke) who were in a clinically verified "stable" phase of recovery was conducted with the PBPT protocol, in which patients made over 12,000 visually guided, point-to-point movements., (cont.) Though prior clinical results suggested that recovery would plateau 6 months post-stroke, two studies using sensorimotor (SM) and progressive resistance (PR) therapy protocols achieved significant, though modest, reductions in impairment. The new PBPT protocol produced significantly larger impairment reductions with over 6,000 fewer movements than SM and PR. By design, the adapting PBPT parameters, namely, the time allotted to move between targets and the virtual slot sidewall stiffness, serve as indicators of patients' abilities to move and aim (as parameters decrease (increase), patients move faster (slower) and require less (more) aiming assistance). By analyzing the parameters' evolution throughout the PBPT protocol, it was shown that motor recovery follows an exponential progression similar to a motor learning "law of practice". In addition, a serendipitous benefit of the PBPT protocol occurred - a sustained reduction in abnormal muscle tone, a factor unrelated to learning. A spectral impedance estimation method suitable for a clinical setting was developed and validated by identifying known mechanical systems. In addition, preliminary data was collected on unimpaired subjects and stroke patients., by Jerome J. Palazzolo., Ph.D.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2004., Includes bibliographical references (p. 165-168)., In this thesis, I explored the design of a thin variable impedance material using electrorheological (ER) fluid that is intended to be worn by humans. To determine the critical design parameters of this material, the shear response of a sandwich of electrodes separated by ER fluid and several different spacer materials was investigated. After a preliminary test to verify that the shear response is controllable by an applied voltage, a single-axis tensile testing machine was designed and constructed to carry out more accurate testing. Two different ER fluids, homogeneous and heterogeneous were investigated. A model of the material for each fluid along with a general model were developed and the parameters of the models were determined through experiments. The model shows a good fit to the experimental data for the heterogeneous fluid based materials, with prediction errors on the order of 30% for two of the spacer materials. The homogeneous fluid based materials show a strong deviation from the model at OV, but fit well when voltage was applied. Polypropylene as a spacer dramatically reduced or eliminated the ER effect. Some critical design parameters identified include: variation in electrode spacing, spacer material selection, and breakdown levels., by Douglas Elmer Eastman, IV., S.M.

Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2003., Includes bibliographical references (p. 209-214)., Human motor control pathologies, such as those caused by stroke, can be treated through physical rehabilitation. The use of robots in therapy environments seems appropriate considering the possibilities they offer for quantification of performance as well as "quality control" between therapy sessions. The research presented in this thesis is part of the continuing development of robotic applications for physical therapy and neuro-rehabilitation at the Newman Laboratory for Biomechanics and Human Rehabilitation. MIT-MANUS, a robot for shoulder and elbow therapy developed in this lab, introduced this new brand of therapy, offering a highly backdrivable mechanism with a soft and stable feel for the user. The focus here is the development, characterization, and implementation of a robot for wrist rehabilitation, designed to provide three rotational degrees of freedom. The wrist motions of flexion/extension and abduction/adduction are governed by a differential gear mechanism, while pronation and supination of the forearm are actuated by a curved slider attached to the rest of the mechanism. Through the characterization, the device was found to exhibit some unwanted behavior, largely attributable to the nonlinearities inherent in the system. Efforts to suppress these effects through control are presented along with recommendations for addressing these problems at the design level., (cont.) The alpha prototype has been set up for clinical trials by providing a functional control scheme along with "video game" patient interfaces; initial clinical trials will run in parallel with the development of the next version of the device. If improvements comparable to those seen with the use of MIT-MANUS are seen with the wrist robot, then rehabilitation therapists will have a new and useful tool at their disposal., by James R. Celestino., S.M.

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Mechanical Engineering, 2005., Includes bibliographical references (p. 347-359)., Robots that engage in significant physical interaction with humans, such as robotic physical therapy aids, must exhibit desired mechanical endpoint impedance while simultaneously producing large forces. In most practical robot configurations, this requires actuators with high force-to-weight ratios and low intrinsic impedance. This thesis explores several approaches to improve the tradeoff between actuator force capacity, weight, and ability to produce desired impedance. Existing actuators that render impedance accurately generally have poor force densities while those with high force densities often have high intrinsic impedance. Aggressive force feedback can reduce apparent endpoint impedance, but compromises coupled stability. The common standard for ensuring coupled stability, passivity, can limit performance severely. An alternative measure of coupled stability is proposed that uses limited knowledge of environment dynamics (e.g. a human limb) and applies robust stability tools to port functions. Because of structural differences between interaction control and servo control, classical single-input, single-output control tools cannot be directly applied for design. Instead, a search method is used to select controller parameters for an assumed structure., (cont.) Simulations and experiments show that this new approach can be used to design a force-feedback controller for a robot actuator that improves performance, reduces conservatism, and maintains coupled stability. Adding dynamics in series to change an actuator's physical behavior can also improve performance. The design tools developed for controller design are adapted to select parameters for physical series dynamics and the control system simultaneously. This design procedure is applied to both spring-damper and inertial series dynamics. Results show that both structures can be advantageous, and that the systematic design of hardware and control together can improve performance dramatically over prior work. A remote transmission design is proposed to reduce actuator weight directly. This design uses a stationary direct-drive electromagnetic actuator and a passive, flexible hydraulic transmission with low intrinsic impedance, thereby utilizing the impedance- rendering capabilities of direct-drive actuation and the force density of hydraulic actuation. The design, construction and characterization of a low-weight, low-friction prototype for a human arm therapy robot are discussed. Recommendations and tradeoffs are presented., by Stephen Paul Buerger., Ph.D.