Research Group Projects and Descriptions

We know from early Roman mosaics that physical rehabilitation and amplification technologies have been used during much of recorded history. Although the goal of constructing such technologies is not new, great scientific and technological hurdles still remain. Even today, permanent assistive devices are viewed by the physically challenged as separate, lifeless mechanisms and not intimate extensions of the human body—structurally, neurologically, and dynamically. The Biomechatronics group seeks to advance technologies that promise to accelerate the merging of body and machine, including device architectures that resemble the body's own musculoskeletal design, actuator technologies that behave like muscle, and control methodologies that exploit principles of biological movement.

We are designing a powered hip-knee-ankle orthosis to assist individuals suffering from neuromuscular disorders. In contrast to mechanically passive, commercially available orthoses, we are adding actuation and control to enhance movement of hip, knee, and ankle joints. We motivate our control scheme with kinematic gait data measured from unimpaired humans walking at slow, moderate, and fast speeds. However, instead of simply tracking biological trajectories, our controller enforces a particular whole-body, angular momentum partition (how the angular momentum of individual body segments contributes to total-body angular momentum). In our current work, orthosis hardware systems are being designed and biomimetic controllers are being evaluated using morphologically realistic, humanoid walking simulations.

We are designing an active ankle-foot orthosis for the treatment of ankle pathologies resulting from disease or traumatic injury. Muscle-like actuators are being advanced to control both joint impedance and motive force during inversion and eversion as well as plantar and dorsi flexion movements. In addition to this work, we are developing global orientation sensors to send ankle angle and ground reaction force measurements from the unaffected leg to the controller on the affected side. Using these sensory signals, the actuators move the affected limb to mimic the unaffected side, thereby restoring gait symmetry.

Transients occur in human walking during a transition to, from, and between steady state walking and acts as an impulse destabilizing a gait cycle. Turns, rapid stops, and accelerated starts are all common transients encountered and managed intelligently by humans everyday. Humanoid bipeds are rapidly becoming a more common part of our everyday life. Therefore, they must also be able to navigate our environments adroitly if they are to assist us in our daily living. This project takes biomechanical principals of angular momentum and applies them to design of controllers for bipeds using angular momentum primitives. These primitives are basic units that simplify the control problem and reduce the dimensionality of the state-space and the objective task. The task in this project is to accomplish transient behaviors and steady state walking together. Through this we are able to realize a more efficient and effective control for humanoid robots.

Alumni Contributor(s): Marko Popovic

Human walking neuromechanical models show how each muscle works during normal, level-ground walking. They are mainly modeled with clutches and linear springs, and are able to capture dominant normal walking behavior. This suggests to us to use a knee-flexor spring for below-knee amputees. We have developed the powered ankle prosthesis, which generates enough force to enable a user to walk "normally." However, amputees still have problems at the knee joint due to the lack of gastrocnemius, which works as an ankle-knee flexor and a plantar flexor. We hypothesize that metabolic cost and EMG patterns of an amputee with our powered ankle and virtual gastrocnemius will dramatically improve.

The majority of commercial prosthetic knees are passive and cannot replicate the positive mechanical work exhibited by the intact human knee in early and late stance. In contrast to traditional, purely dissipative prosthetic knees, we propose a biomimetic knee, with antagonistic actuation, designed to reproduce both positive and negative work phases of the intact joint while using series elasticity to minimize net energy consumption. We present the design and physical implementation of the active knee prosthesis prototype. The objective of this adaptive powered prosthetic device is to improve gait and metabolic energy consumption of above-knee amputees, on variant terrain conditions.

We are developing a control architecture for bipedal locomotion devices such as robots and powered orthotics. Advanced nonlinear control techniques, including feedback linearization, sliding control, and multivariable optimization, are utilized in this control architecture, yielding a highly stable and tunable controller for a highly unstable and nonlinear plant. Tests—using a 3-D, 12-degree-of-freedom humanoid model—include a variety of disturbed initial states and of control goals for the center of mass, swing foot, and other points being controlled. An interesting property of this controller is the emergence of appropriate non-contact limb behavior in response to disturbances. Also, due to its large range of operation, this control architecture can reject significant disturbances more easily than simpler controllers, and requires a less-detailed reference trajectory than simpler controllers. This has the additional benefit of reducing the computational workload of a motion planner in an integrated motion planning and control system. Such control architectures will find use in assistive devices for the elderly and handicapped.

Alumni Contributor(s): Marko Popovic

Motivated by applications in rehabilitation and robotics, we are developing methodologies to control muscle-actuated systems via electrical stimulation. As a demonstration of such potential, we are developing centimeter-scale robotic systems that utilize muscle for actuation and glucose as a primary source of fuel. This is an interesting control problem because muscles: a) are mechanical state-dependent actuators; b) exhibit strong nonlinearities; and c) have slow time-varying properties due to fatigue-recuperation, growth-atrophy, and damage-healing cycles. We are investigating a variety of adaptive and robust control techniques to enable us to achieve trajectory tracking, as well as mechanical power-output control under sustained oscillatory conditions. To implement and test our algorithms, we developed an experimental capability that allows us to characterize and control muscle in real time, while imposing a wide variety of dynamical boundary conditions.

We study the mechanical behavior of leg muscles and tendons during human walking in order to motivate the design of economical robotic legs. We hypothesize that quasi-passive, series-elastic clutch units spanning the knee joint in a musculoskeletal arrangement can capture the dominant mechanical behaviors of the human knee in level-ground walking. Biarticular elements necessarily need to transfer energy from the knee joint to hip and/or ankle joints, and this mechanism would reduce the necessary muscle work and improve the mechanical economy of a human-like walking robot.

This research aims to extract a potentially small set of underlying principles that govern human movement and to apply that set of principles to biomimetic control systems. Using a morphologically realistic human model and kinematic gait data, we find that spin angular momentum in human walking is highly regulated, and that there exists a nonlinear coupling between center of mass transverse forces, center of mass position, and center of pressure location. Using an open loop optimization strategy, we show that biologically realistic leg joint kinematics emerge through the minimization of spin angular momentum and the sum of the joint torques squared. This suggests that both angular momentum and energetic factors are important considerations for biomimetic controllers.

Towards the goal of developing stable humanoid robots and leg prostheses/othoses, we are developing a biologically motivated control system for walking where system angular momentum is explicitly controlled. Using human kinematic walking data, we find that spin angular momentum is highly regulated in walking. In addition, our analysis shows that the distribution of angular momentum throughout the human body, or the angular momentum partition, is invariant with walking speed. Motivated by these biomechanical results, we conduct numerical simulations of walking using a morphologically realistic human model. Our control system searches for joint reference trajectories that minimize the error between the model's angular momentum partition and the biologically determined partition. In order to understand motor control in humans, we are experimenting with biological time delays (order 100 ms) and correlating our simulation results with human behavior.

Alumni Contributor(s): Marko Popovic

Developing biologically inspired robotic prostheses necessitates precise understanding of the dynamic interaction between amputee, prosthesis, and the environment they act on. In our research, we are instrumenting a biomimetic ankle-foot powered prosthesis prototype with a series of sensory units to estimate the ground reaction forces (GRF) and zero moment point (ZMP) trajectory. The incorporation of this sensory information with a morphologically realistic human model and basic feedback methods will contribute to the development of balance-control strategies in the mentioned device. These strategies will enhance amputees’ perception and control of their dynamic stability. With this new generation of robotic ankle-foot prostheses, we are addressing some of the main difficulties that amputees encounter with current passive devices, including non-symmetric gait, increased walking energy cost, and appropriate maintenance of balance during standing and walking.

We are developing a robust, low-power, stable, and lightweight leg orthosis designed to dramatically increase the locomotory endurance of humans. Instead of the biological leg actively stiffening to support body weight throughout each locomotory step, the orthosis, running parallel to the leg, will offer that support, deactivating leg muscles and dramatically lowering metabolic demands. The proposed transportation technology promises to dramatically reduce the human effort required to traverse rough terrains in urban and natural environments.

Alumni Contributor(s): Andrew Valiente

Augmentation of human locomotion has proven an elusive goal. Natural human walking is extremely efficient and the complex articulation of the human leg poses significant engineering difficulties. We present a wearable exoskeleton designed to reduce the metabolic cost of jogging. The exoskeleton places a stiff fiberglass spring in series with the complete leg during stance phase, then removes it so that the knee may bend during leg swing. The result is a bouncing gait with reduced reliance on the musculature of the knee and ankle.

A large percentage of the work done at the knee during normal walking is dissipative. A rotary magnetorheological damper along the knee axis can exert high damping forces with only minimal electrical power. This method has distinct advantages over conventional, fully passive, and modern hydraulic transfemoral prostheses. Biologically realistic control of the MR prosthesis can reduce the metabolic rate of the user and improve biological cosmesis.

People on the autism spectrum face a number of challenges, including motor movement issues that can cause limbs to cease activity. Circumstantial evidence suggests that autonomic nervous system influences related to stress and overload may arise from and contribute to these problems. We propose to allow individuals to monitor several physiological parameters to see if there are patterns that recognize or predict the onset of their individual motor problems. We plan to develop new, wearable technology to treat these problems via the use of tiny, vibrotactile devices carefully placed at the joints. We hypothesize that some methods of touch-feedback and vibration at the joints may enable individuals to recover motor functioning during episodes of intermittent loss. We are also exploring the development of personally controlled devices that facilitate finer motor movement for augmenting communication as needed for assisting in typing or pointing.

The human ankle provides a significant amount of net positive work during the stance period of walking, especially at moderate to fast walking speeds. On the contrary, conventional ankle-foot prostheses are completely passive during stance, and consequently, cannot provide net positive work. Clinical studies indicate that transtibial amputees using conventional prostheses experience many problems during locomotion including a high gait metabolism, a low gait speed, and gait asymmetry. Researchers believe the main cause for the observed locomotion is due to the inability of conventional prostheses to provide net positive work during stance. The objective of this project is to develop a powered ankle-foot prosthesis that is capable of providing net positive work during the stance period of walking. To this end, we investigate the mechanical design and control system architectures for the prosthesis. We also conduct a clinical evaluation of the proposed prosthesis on different amputee participants.

Difficulties running and participating in active sporting events have long been cited by lower-extremity amputees as among the greatest weaknesses of their prostheses. As a result, several prosthetic feet based on springs have been introduced which enable their users to run and jump. Being passive devices, these prostheses are unable to generate more energy than they absorb, as the natural ankle does, placing an undue burden on the hips and the natural leg. We seek to develop an active prosthetic foot which adapts to its user's running gait and delivers appropriate additional energy. As a result, the foot serves to restore a more natural and efficient running gait. Furthermore, the foot may be used for basic biomechanic research, as it permits the regulation of energy injected into the gait cycle.

Muscles are biological actuators with unique properties compared to traditional mechanical actuators. The ability to simultaneously modify stiffness while providing power as well as the potential for self-repair make their use desirable when considering the design of robotic systems. Improving our understanding of how muscles work as actuators, struts, and springs is essential for both robotic design and for understanding animal locomotion. Unfortunately, our understanding of the energetics of muscles is incomplete, and traditional methods for studying muscles have inherent limitations. Our group has developed a novel apparatus that allows us to test muscles in unique ways by tethering the muscle to a movable platform coupled to a computer-simulated load. With this setup, we can vary the parameters of the "virtual load" to address more complex and relevant questions, and examine the dynamic interactions between muscles.

Alumni Contributor(s): Waleed A. Farahat