Thomas Schauer

IMU-based joint angle measurement for gait analysis.

This contribution is concerned with joint angle calculation based on inertial measurement data in the context of human motion analysis. Unlike most robotic devices, the human body lacks even surfaces and right angles. Therefore, we focus on methods that avoid assuming certain orientations in which the sensors are mounted with respect to the body segments. After a review of available methods that may cope with this challenge, we present a set of new methods for: (1) joint axis and position identification; and (2) flexion/extension joint angle measurement. In particular, we propose methods that use only gyroscopes and accelerometers and, therefore, do not rely on a homogeneous magnetic field. We provide results from gait trials of a transfemoral amputee in which we compare the inertial measurement unit (IMU)-based methods to an optical 3D motion capture system. Unlike most authors, we place the optical markers on anatomical landmarks instead of attaching them to the IMUs. Root mean square errors of the knee flexion/extension angles are found to be less than 1° on the prosthesis and about 3° on the human leg. For the plantar/dorsiflexion of the ankle, both deviations are about 1°.

Control strategies for integration of electric motor assist and functional electrical stimulation in paraplegic cycling: utility for exercise testing and mobile cycling

Aim: The aim of this study was to investigate feedback control strategies for integration of electric motor assist and functional electrical stimulation (FES) for paraplegic cycling, with particular focus on development of a testbed for exercise testing in FES cycling, in which both cycling cadence and workrate are simultaneously well controlled and contemporary physiological measures of exercise performance derived. A second aim was to investigate the possible benefits of the approach for mobile, recreational cycling. Methods: A recumbent tricycle with an auxiliary electric motor is used, which is adapted for paraplegic users, and instrumented for stimulation control. We propose a novel integrated control strategy which simultaneously provides feedback control of leg power output (via automatic adjustment of stimulation intensity) and cycling cadence (via electric motor control). Both loops are designed using system identification and analytical (model-based) feedback design methods. Ventilatory and pulmonary gas exchange response profiles are derived using a portable system for real-time breath-by-breath acquisition. Results: We provide indicative results from one paraplegic subject in which a series of feedback-control tests illustrate accurate control of cycling cadence, leg power control, and external disturbance rejection. We also provide physiological response profiles from a submaximal exercise step test and a maximal incremental exercise test, as facilitated by the control strategy. Conclusion: The integrated control strategy is effective in facilitating exercise testing under conditions of well-controlled cadence and power output. Our control approach significantly extends the achievable workrate range and enhances exercise-test sensitivity for FES cycling, thus allowing a more stringent characterization of physiological response profiles and estimation of key parameters of aerobic function. We further conclude that the control approach can significantly improve the overall performance of mobile recreational cycling.

Data-driven control design for neuroprotheses: a virtual reference feedback tuning (VRFT) approach

This paper deals with design of feedback controllers for knee joint movement of paraplegics using functional electrical stimulation (FES) of the paralyzed quadriceps muscle group. The controller design approach, virtual reference feedback tuning (VRFT), is directly based on open loop measured data and fits the controller in such a way that the closed-loop meets a model reference objective. The use of this strategy, avoiding the modeling step, significantly reduces the time required for controller design and considerably simplifies the rehabilitation protocols. Linear and nonlinear controllers have been designed and experimentally tested, preliminarily on a healthy subject and finally on a paraplegic patient. Linear controller is effective when applied on small range of knee joint angle. The design of a nonlinear controller allows better performances. It is also shown that the control design is effective in tracking assigned knee angle trajectories and rejecting disturbances.

Iterative Learning Control for Variable Pass Length Systems

Abstract Monotonic convergence of linear iterative learning control (ILC) systems with changing pass length is considered. The maximum pass length (MPL) error is introduced as a useful concept for convergence analysis of this class of systems. Using the lifted-system framework, a both necessary and sufficient monotonic convergence criterion is found for the 1-norm of the MPL error. Further findings on 2-norm and ∞-norm convergence are added. Finally, an example system is given, i.e. the control of an electrical stimulation system for gait assistance, and simulation results are provided.

Design of a Symmetry Controller for Cycling Induced by Electrical Stimulation: Preliminary Results on Post-Acute Stroke Patients

This study deals with the design of a controller for cycling induced by functional electrical stimulation. The controller will be exploitable in the rehabilitation of hemiparetic patients who need to recover motor symmetry. It uses the pulse width as the control variable in the stimulation of the two legs in order to nullify the unbalance between the torques produced at the two crank arms. It was validated by means of isokinetic trials performed both by healthy subjects and stroke patients. The results showed that the controller was able to reach, and then maintain, a symmetrical pedaling. In the future, the controller will be validated on a larger number of stroke patients.

A myocontrolled neuroprosthesis integrated with a passive exoskeleton to support upper limb activities

This work aimed at designing a myocontrolled arm neuroprosthesis for both assistive and rehabilitative purposes. The performance of an adaptive linear prediction filter and a high-pass filter to estimate the volitional EMG was evaluated on healthy subjects (N=10) and neurological patients (N=8) during dynamic hybrid biceps contractions. A significant effect of filter (p=0.017 for healthy; p<0.001 for patients) was obtained. The post hoc analysis revealed that for both groups only the adaptive filter was able to reliably detect the presence of a small volitional contribution. An on/off non-linear controller integrated with an exoskeleton for weight support was developed. The controller allowed the patient to activate/deactivate the stimulation intensity based on the residual EMG estimated by the adaptive filter. Two healthy subjects and 3 people with Spinal Cord Injury were asked to flex the elbow while tracking a trapezoidal target with and without myocontrolled-NMES support. Both healthy subjects and patients easily understood how to use the controller in a single session. Two patients reduced their tracking error by more than 60% with NMES support, while the last patient obtained a tracking error always comparable to the healthy subjects performance (<4°). This study proposes a reliable and feasible solution to combine NMES with voluntary effort.

Monotonic convergence of iterative learning control systems with variable pass length

ABSTRACT A growing number of researchers consider iterative learning control (ILC) a promising tool for numerous control problems in biomedical application systems. We will briefly discuss why classical ILC theory is technically too restrictive for some of these applications. Subsequently, we will extend the classical ILC design in the lifted systems framework to the class of repetitive trajectory tracking tasks with variable pass length. We will analyse the closed-loop dynamics for two standard learning laws, and we will discuss in which sense the tracking error can be reduced by which controller design strategies. Necessary and sufficient conditions for monotonic convergence will be derived. We then summarise all results in a set of practical controller design guidelines. Finally, a simulation study is presented, which demonstrates the usefulness of these guidelines and illustrates the special dynamics that occur in variable pass length learning.

Feedback control of arm movements using Neuro-Muscular Electrical Stimulation (NMES) combined with a lockable, passive exoskeleton for gravity compensation.

Within the European project MUNDUS, an assistive framework was developed for the support of arm and hand functions during daily life activities in severely impaired people. This contribution aims at designing a feedback control system for Neuro-Muscular Electrical Stimulation (NMES) to enable reaching functions in people with no residual voluntary control of the arm and shoulder due to high level spinal cord injury. NMES is applied to the deltoids and the biceps muscles and integrated with a three degrees of freedom (DoFs) passive exoskeleton, which partially compensates gravitational forces and allows to lock each DOF. The user is able to choose the target hand position and to trigger actions using an eyetracker system. The target position is selected by using the eyetracker and determined by a marker-based tracking system using Microsoft Kinect. A central controller, i.e., a finite state machine, issues a sequence of basic movement commands to the real-time arm controller. The NMES control algorithm sequentially controls each joint angle while locking the other DoFs. Daily activities, such as drinking, brushing hair, pushing an alarm button, etc., can be supported by the system. The robust and easily tunable control approach was evaluated with five healthy subjects during a drinking task. Subjects were asked to remain passive and to allow NMES to induce the movements. In all of them, the controller was able to perform the task, and a mean hand positioning error of less than five centimeters was achieved. The average total time duration for moving the hand from a rest position to a drinking cup, for moving the cup to the mouth and back, and for finally returning the arm to the rest position was 71 s.

Automatic Control of a Drop‐Foot Stimulator Based on Angle Measurement Using Bioimpedance

The topic of this contribution is iterative learning control of a drop-foot stimulator in which a predefined angle profile during the swing phase is realized. Ineffective dorsiflexion is compensated by feedback-controlled stimulation of the muscle tibialis anterior. The ankle joint measurement is based on changes in the bioimpedance (BI) caused by leg movements. A customized four-channel BI measurement system was developed. The suggested control approach and the new measurement method for the joint angle were successfully tested in preliminary experiments with a neurologically intact subject. Reference angle measurements were taken with a marker-based optical system. An almost linear relation between joint angle and BI was found for the angle range applicable during gait. The desired angle trajectory was closely tracked by the iterative learning controller after three gait cycles. The final root mean square tracking error was below 5 degrees.

Alignment-Free, Self-Calibrating Elbow Angles Measurement Using Inertial Sensors

Due to their relative ease of handling and low-cost, inertial measurement unit (IMU) based joint angle measurements are used for a widespread range of applications. These include sports performance, gait analysis and rehabilitation (e.g. Parkinsons disease monitoring or post-stroke assessment). However, a major downside of current algorithms recomposing human kinematics from IMU data is that they require calibration motions and/or the careful alignment of the IMUs respective to their body segment. In this article, we propose a new method, which is alignment free and self-calibrated using the arbitrary movements of the user and an initial zero reference arm pose. The proposed method utilizes real time optimization to identify the two dominant axes of rotation of the elbow joint. Using a two degree of freedom joint mimicking the human elbow, the performance of the algorithm was assessed by comparing the angles obtained from two IMUs to the ones obtained from a marker-based optical tracking system. The self-calibration proved to converge within seconds and the RMS errors with respect to the optical reference system were below 5°. Our method can be particularly useful in the field of telerehabilitation, where precise manual sensor to segment alignment as well as precise, predefined calibration movements are impractical.