Fiabilidad de la cinemática angular de rodilla y tobillo de la carrera en el plano sagital medido con Younext Direct Motion Capture®

En los últimos cuarenta años la popularidad de salir a correr o hacer running se ha incrementado notablemente. Su popularidad y el gran número de lesiones hace que el running sea la actividad física más estudiada por el área de la biomecánica. En la práctica clínica, se utilizan principalmente software de análisis de movimiento 2D para analizar la cinemática. Younext Direct Motion Capture® (Podoactiva S.L., Huesca) es un software que analiza los parámetros cinemáticos durante la carrera en el plano sagital. El objetivo general de este estudio fue determinar la fiabilidad y repetibilidad de la cinemática angular de la rodilla y del tobillo de la carrera en el plano sagital en población sana utilizando Younext Direct Motion Capture®. Estudio piloto observacional de concordancia. 8 sujetos (5 varones y 3 mujeres) fueron filmados mientras corrían en una cinta de correr, obteniendo tres grabaciones del ciclo de la carrera de cada participante en la primera sesión. Se realizó una segunda sesión de grabaciones a los 7 días del primer registro para volver a grabar siguiendo el mismo protocolo. Se calculó el coeficiente de correlación intraclase (ICC) para determinar la fiabilidad intra-sesión e inter-sesión del software a la hora de medir la cinemática de la rodilla y del tobillo. Se obtuvo una fiabilidad intra-sesión e inter-sesión casi perfectas (ICC> 0,81) tanto para tobillo como para rodilla. Este estudio demostró que el software Younext Direct Motion Capture® es fiable para analizar la cinemática angular de la rodilla y del tobillo de la carrera en población adulta sana.

Model-Free Adaptive Control of Direct Drive Servo Valve of Electromagnetic Linear Actuator

An electromagnetic linear actuator (EMLA) has a promising application in direct motion control. However, ELMA will inevitably inherit uncertainties in the face of load changes, system parameter perturbation, and inherent system nonlinearities, all of which constitute disturbances adversely affecting the precision and adaptability of the control system. A model-free adaptive control (MFAC) strategy based on full form dynamic linearization (FFDL) was proposed to reduce the sensitivity of the control system to the disturbances. An adaptive control of direct drive servo valve was achieved based on the online interaction of characteristic parameters and control algorithms. The feasibility and precision of the proposed algorithm were verified through simulation and experimental results. The results show that the proposed algorithm could achieve adaptive adjustment of the servo valve response at different openings of 0-3 mm without changing control parameters, with the response time controlled within 10ms and steady state error less than 0.04mm. Furthermore, the proposed algorithm had better robustness and capacity of resisting disturbance.

Preventing Tumbling With a Twisting Trunk for the Quadruped Robot: Origaker I

The tumble stability indicates the capability to resist the tumble caused by disturbances. For a quadruped robot, the tumble is mainly about the line segment connecting two supporting feet. The tumble stability of quadruped robots is evaluated by various stability criteria based on forces, moments or energies. Work has been done to improve the tumble stability of quadruped robots. Nevertheless, the previous work to achieve this goal relied on motion of legs. No trunk motions were considered. As a matter of fact, trunk motion is widely utilized by natural quadrupeds. By utilizing trunk motion, the quadrupeds are able to regulate the center of gravity to improve the tumble stability level. This paper for the first time investigates the effect of the twisting trunk on the tumble stability of quadruped robots from the viewpoint of energy. Thus it can be seen that the twisting trunk help improve the tumble stability level of quadruped robots. The relationship between the tumble stability and trunk twisting is to be analyzed mathematically, and help find the maximum disturbing energy that the quadruped robot can bear with a twisting trunk and further direct motion of the trunk twisting during tumbles to prevent any overturning.

Admissible Abstractions for Near-optimal Task and Motion Planning

We define an admissibility condition for abstractions expressed using angelic semantics and show that these conditions allow us to accelerate planning while preserving the ability to find the optimal motion plan. We then derive admissible abstractions for two motion planning domains with continuous state. We extract upper and lower bounds on the cost of concrete motion plans using local metric and topological properties of the problem domain. These bounds guide the search for a plan while maintaining performance guarantees. We show that abstraction can dramatically reduce the complexity of search relative to a direct motion planner. Using our abstractions, we find near-optimal motion plans in planning problems involving 10^13 states without using a separate task planner.