Jochen Schuy

Power-optimized stiffness and nonlinear position control of an actuator with Variable Torsion Stiffness

Introducing compliant actuation to robotic joints is an approach to ensure safety in closer human-machine interaction. Further, the possibility to adjust stiffness can be beneficial considering energy storage and the power consumption required to track certain trajectories. The subject of this paper is the stiffness and position control of the Variable Torsion Stiffness (VTS) actuator for application in compliant robotic joints. For the realization of a variable rotational stiffness, the active length of a torsional elastic element in serial configuration between drive and link is adjusted in VTS. After the deduction of an extended drive train model, this paper gives an advanced power analysis clarifying power-optimal settings from previous basic models and identifying additional settings that allow for a more versatile operation. Based on these results that can be generalized to other variable elastic actuator concepts, an optimized strategy for setting stiffness is determined considering the whole system dynamics including natural frequencies as well as antiresonance effects. For position control of VTS in a prototypical implementation, a nonlinear position controller is designed by means of feedback linearization. Although the system is modified significantly by changing drive train stiffness, the stiffness adaptation of the controller ensures the required tracking performance.

Design and control of a robot for the assessment of psychological factors in prosthetic development

This paper introduces a robotic concept for the assessment of psychological factors in prosthetic design. Its aim is to imitate the postural movements of the participants while those are conducting squatting movements in order to investigate the integration of artificial limbs to the subjects body scheme. Therefore, the robot mimics the functionality and appearance of the human foot, shank and thigh as well as the ankle and knee joint. To induce a more realistic outer appearance, the hull of a shop-window mannequin is used as cladding. The robot is controlled by a computed torque control combined with a RGB-D sensor for the acquisition of the desired trajectories from the participant. In the test setup one leg of the participant is hidden from his view while the robot stands next to him and imitates the movements of this leg. This paper gives an insight in the theory of body schema integration. The concept of the robot is described and detailed information about the mechanical design and actuator dimensioning in accordance with psychological and biomechanical requirements are given. Furthermore, the concept of the human-machine interface, the control algorithm and simulations based on experimental data from a human subject are presented.

Dimensioning and evaluation of the elastic element in a Variable Torsion Stiffness actuator

Variable stiffness joints are essential for robots or biomechanical applications to generate torque with sufficient compliance to ensure safety simultaneously. This paper presents the dimensioning of the elastic element for an actuator with Variable Torsion Stiffness (VTS). Analytical and finite element calculations are compared to experimental evaluation of a splined shaft profile realized in a prototype implementation. Based on simplifications and effects in the real test-rig the results show similar behaviour. Furthermore, an analytical and finite element investigation of four cross-sections with torsional load illustrate optimal utilization of cylindrical geometries considering the torsional stress. In contrast, the low stressed edges of uncylindrical cross-sections are adequate areas to bear additional stresses.

Integrated measurement system for amputee gait analysis: A pilot study

This paper presents a new direct measurement system which enable an amputee gait assessment independent of location, not restricted in ground conditions or tasks. As an adapter it is implemented in the structure of the lower limb prosthesis to measure load and motions in six degrees of freedom. A pilot study of typical tasks for gait analysis applying the integrated measurement system is shown. The resulting measured parameter of the system demonstrates reasonable data compared to reference data of able-bodied humans.

Measurement of biomechanical interactions at the stump-socket interface in lower limb prostheses

This paper introduces a novel measuring approach for detecting relative movement between stump and socket in lower limb prostheses. The application of the motion capturing based measuring approach is shown at a single male trans-tibial amputee using a Patella Tendon Bearing (PTB) socket. It further investigates and assesses the feasibility of measuring the relative movement between stump and socket during level walking at different velocities and allocating it to the coinciding loads. Representative results for the two translational degrees of freedom in the sagittal plane are presented and discussed. For the proximodistal (pd) direction, a linear correlation between applied load and relative movement is found, while for the anteroposterior (ap) direction the stump movement is largely influenced by the motion sequence during the respective gait event. Additionally, the effect of walking speed is discussed.

A new device to measure load and motion in lower limb prosthesis — Tested on different prosthetic feet

In this work, a new system is presented to measure load and motion in six degrees of freedom synchronously to assess amputee gait in real time. The device is integrated in the shank of lower limb prostheses to measure the gait parameters based on a combination of AHRS (Attitude Heading Reference System) and a strain gauge pylon. This novel mobile and autonomous system enables new possibilities of amputee gait assessment by gathering both kinds of data. After a presentation of the system, two approaches to calculate the 6×6 calibration matrix of the strain gauge pylon with respect to main channels and cross talk channels are presented and compared. Finally, the data analysis process, a performed pilot study of a transfemoral amputee wearing two different prosthetic feet and consequent load results are discussed.

A User-Specific Human-Machine Interaction Strategy for a Prosthetic Shank Adapter

Abstract For people with lower limb amputation, a user-specific human-machine interaction with their prostheses is required to ensure safe and comfortable assistance. Especially during dynamic turning manoeuvres, users experience high loads at the stump, which decreases comfort and may lead to long-term tissue damage. Preliminary experiments with users wearing a configurable, passive torsional adaptor indicate increased comfort and safety achieved by adaptation of torsional stiffness and foot alignment. Moreover, the results show that the individual preference regarding both parameters depend on gait situation and individual preference. Hence, measured loads in the structure of the prosthesis and subjective feedback regarding comfort and safety during different turning motions are considered in a user-specific human-machine interaction strategy for a prosthetic shank adaptor. Therefore, the interrelations of gait parameters with optimal configuration are stored in an individual preference-setting matrix. Stiffness and foot alignment are actively adjusted to the optimal parameters by a parallel elastic actuator. Two subjects reported that they experienced appropriate variation of stiffness and foot alignment, a noticeable reduction of load at the stump and that they could turn with less effort.

Design & evaluation of a sensor minimal gait phase and situation detection Algorithm of Human Walking

This paper presents the design and evaluation of gait detection algorithm based on one IMU placed on the shank. The algorithm is based on adaptive thresholds by artificial neural network and fuzzy logic to identify gait phase and situation for real-time applications like micro-processed prosthesis. Offline evaluation with fifteen able-bodied subjects and two transtibial amputees shows high detection rates of 98 % for distinguishing stance from swing phase as well as 93.6 % between straight and turning gait situation with global parameters.