Zeitschriftenaufsätze und Buchbeiträge

Dogs (Canis familiaris) prefer the walk at lower speeds and the more economical trot at speeds ranging from 0.5 Fr up to 3 Fr. Important works have helped to understand these gaits at the levels of the center of mass, joint mechanics, and muscular control. However, less is known about the global dynamics for limbs and if these are gait or breed-specific. For walk and trot, we analyzed dogs’ global dynamics, based on motion capture and single leg kinetic data, recorded from treadmill locomotion of French Bulldog (N = 4), Whippet (N = 5), Malinois (N = 4), and Beagle (N = 5). Dogs’ pelvic and thoracic axial leg functions combined compliance with leg lengthening. Thoracic limbs were stiffer than the pelvic limbs and absorbed energy in the scapulothoracic joint. Dogs’ ground reaction forces (GRF) formed two virtual pivot points (VPP) during walk and trot each. One emerged for the thoracic (fore) limbs (VPPTL) and is roughly located above and caudally to the scapulothoracic joint. The second is located roughly above and cranially to the hip joint (VPPPL). The positions of VPPs and the patterns of the limbs’ axial and tangential projections of the GRF were gaits but not always breeds-related. When they existed, breed-related changes were mainly exposed by the French Bulldog. During trot, positions of the VPPs tended to be closer to the hip joint or the scapulothoracic joint, and variability between and within breeds lessened compared to walk. In some dogs, VPPPL was located below the pelvis during trot. Further analyses revealed that leg length and not breed may better explain differences in the vertical position of VPPTL or the horizontal position of VPPPL. The vertical position of VPPPL was only influenced by gait, while the horizontal position of VPPTL was not breed or gait-related. Accordingly, torque profiles in the scapulothoracic joint were likely between breeds while hip torque profiles were size-related. In dogs, gait and leg length are likely the main VPPs positions’ predictors. Thus, variations of VPP positions may follow a reduction of limb work. Stability issues need to be addressed in further studies.

The parallel connection of technical and biological systems with a comparable mechanical behavior offers the possibility of reducing the interaction forces between those systems. Especially in the context of human-robot interaction (e.g., exoskeletons), it can improve user safety and acceptance at the same time. With this aim, we used antagonistic actuators with nonlinear compliance for a modular upper-extremity exoskeleton following biological paragons, mirroring the “blueprint” of its human user. In a test-bed setup, we compared antagonistic compliant actuation with antagonistic stiff, unilateral stiff and unilateral compliant actuation in the artificial “elbow joint” of the exoskeleton test bed. We show that this type of actuation allows the variation of the joint stiffness during motion, independent of the position. With the approach we propose, compliance leads to reduced force peaks and angular jerk, without sacrifices in terms of time constants and overshoot of amplitudes. We conclude that the presented actuation principle has considerable benefits in comparison to other types of exoskeleton actuation, even when using only commercially available and 3D printed components. Based on our work, further investigations into the control of compliant antagonistically actuated exoskeletons become realizable.

Biomechatronics is an engineering subject in which biomimetics as a method is one of its two supporting pillars: biology for engineering, or Bio4Eng. This is contrasted with biocompatible design, or Eng4Bio, examples of which are human-serving systems, such as exoskeletons, and biomedical engineering. The paper aims to illustrate that the research fields of biomimetics, biomechatronics, and biomedical engineering are not in competition but mutually supportive. The current attempts to place biomechatronics under the umbrella of biomimetics or biomedical engineering are therefore not expedient; they deprive the subject of its strength of combining Bio4Eng and Eng4Bio at any time in a task-related manner. In addition to research and development, however, the training of the specialists supporting the subjects must not be disregarded and is therefore described based on a proven design.