This paper presents a personalized gait optimization framework for lower-body exoskeletons. Rather than optimizing numerical objectives such as the mechanical cost of transport, our approach directly learns from user preferences, e.g., for comfort. Building upon work in preferencebased interactive learning, we present the COSPAR algorithm. COSPAR prompts the user to give pairwise preferences between trials and suggest improvements; as exoskeleton walking is a non-intuitive behavior, users can provide preferences more easily and reliably than numerical feedback. We show that COSPAR performs competitively in simulation and demonstrate a prototype implementation of COSPAR on a lower-body exoskeleton to optimize human walking trajectory features. In the experiments, COSPAR consistently found user-preferred parameters of the exoskeleton's walking gait, which suggests that it is a promising starting point for adapting and personalizing exoskeletons (or other assistive devices) to individual users.

With the goal of moving towards implementation of increasingly dynamic behaviors on underactuated systems, this paper presents an optimization-based approach for solving full-body dynamics based controllers on underactuated bipedal robots. The primary focus of this paper is on the development of an alternative approach to the implementation of controllers utilizing control Lyapunov function based quadratic programs. This approach utilizes many of the desirable aspects from successful inverse dynamics based controllers in the literature, while also incorporating a variant of control Lyapunov functions that renders better convergence in the context of tracking outputs. The principal benefits of this formulation include a greater ability to add costs which regulate the resulting behavior of the robot, in addition, the model error-prone inertia matrix is used only once, in a non-inverted form. The result is a successful demonstration of the controller for walking in simulation, and applied on hardware in real-time for crouching.

Robotic devices have been proposed to meet the rising need for high intensity, long duration, and goal-oriented therapy required to regain motor function after neurological injury. Complementing this application, exoskeletons can augment traditional clinical assessments through precise, repeatable measurements of joint angles and movement quality. These measures assume that exoskeletons are making accurate joint measurements with a negligible effect on movement. For the coupled and coordinated joints of the wrist and hand, the validity of these two assumptions cannot be established by characterizing the device in isolation. To examine these assumptions, we conducted three user-in-the-loop experiments with able-bodied participants. First, we compared robotic measurements to an accepted modality to determine the validity of joint- and trajectory-level measurements. Then, we compared those movements to movements without the device to investigate the effects of device dynamic properties on wrist movement characteristics. Last, we investigated the effect of the device on coordination with a redundant, coordinated pointing task with the wrist and hand. For all experiments, smoothness characteristics were preserved in the robotic kinematic measurement and only marginally impacted by robot dynamics, validating the exoskeletons for use as assessment devices. Stemming from these results, we propose design guidelines for exoskeletal assessment devices.

No abstract

Abstract-Robotic exoskeletons can provide the high intensity, long duration targeted therapeutic interventions required for regaining motor function lost as a result of neurological injury. Quantitative measurements by exoskeletons have been proposed as measures of rehabilitative outcomes. Exoskeletons, in contrast to end effector designs, have the potential to provide a direct mapping between human and robot joints. This mapping rests on the assumption that anatomical axes and robot axes are aligned well, and that movement within the exoskeleton is negligible. These assumptions hold well for simple one degree-of-freedom joints, but may not be valid for multiarticular joints with unique musculoskeletal properties such as the wrist. This paper presents an experiment comparing robot joint kinematic measurements from an exoskeleton to anatomical joint angles measured with a motion capture system. Joint-space position measurements and task-space smoothness metrics were compared between the two measurement modalities. The experimental results quantify the error between joint-level position measurements, and show that exoskeleton kinematic measurements preserve smoothness characteristics found in anatomical measures of wrist movements.

Collective opinions affect civic participation, governance, and societal norms. Due to the influence of opinion dynamics, many models of their formation and evolution have been developed. A commonly used approach for the study of opinion dynamics is bounded-confidence models. In these models, individuals are influenced by the opinions of others in their network. They generally assume that individuals will formulate their opinions to resemble those of their peers. In this paper, inspired by the dynamics of partisan politics, we introduce a bounded-confidence model in which individuals may be repelled by the opinions of their peers rather than only attracted to them. We prove convergence properties of our model and perform simulations to study the behavior of our model on various types of random networks. In particular, we observe that including opinion repulsion leads to a higher degree of opinion fragmentation than in standard bounded-confidence models.

With the goal of moving towards implementation of increasingly dynamic behaviors on underactuated systems, this paper presents an optimization-based approach for solving full-body dynamics based controllers on underactuated bipedal robots. The primary focus of this paper is on the development of an alternative approach to the implementation of controllers utilizing control Lyapunov function based quadratic programs. This approach utilizes many of the desirable aspects from successful inverse dynamics based controllers in the literature, while also incorporating a variant of control Lyapunov functions that renders better convergence in the context of tracking outputs. The principal benefits of this formulation include a greater ability to add costs which regulate the resulting behavior of the robot, in addition, the model error-prone inertia matrix is used only once, in a non-inverted form. The result is a successful demonstration of the controller for walking in simulation, and applied on hardware in real-time for crouching.