Logo des Repositoriums

Auflistung nach Autor:in "Ji, Linhong"

1 - 2 von 2
  • Konferenzbeitrag
    Intelligent control based on location-force model
    (Informatik 2014, 2014) Sui, Jianfeng; Ji, Linhong
    This article introduces a multi-posture locomotor training device (MPLTD)with a closed-loop control scheme based on joint-angle feedback, which is able to overcome varies difficulties resulting from mechanical vibration and the weight of trainer to achieve higher accuracy trajectory. By introducing the forcefield control scheme that used in the closed-loop control, the device can obtain the active-constrained mode including the passive one. The MPLTD is mainly composed of three systems: posture adjusting and weight support system, lower limb exoskeleton system, and control system, of which the lower limb exoskeleton system mainly includes the indifferent equilibrium mechanism with two degrees of freedom (DOF) and the driving torque is calculated by the Lagrangian function. In addition, a series of experiments-the weight support and the trajectory accuracy experiment demonstrate a good performance of mechanical structure and the closed-loop control.
  • Konferenzbeitrag
    Optimizing patients' engagement by a cyber-physical rehabilitation system
    (Informatik 2014, 2014) Li, Chong; Rusák, Zoltán; Horváth, Imre; Hou, Yuemin; Ji, Linhong
    Successful rehabilitation of stroke patients is strongly dependent on the engagement of patients. During a whole rehabilitation program, mundane rehabilitation exercises can easily become routine for patients, leading to boredom and as a result to ineffective functional recovery. This has been taken into consideration just by few rehabilitation systems. Engagement in rehabilitation can be decomposed into long term engagement (LTE) and short term engagement (STE) for the reason that their indicators and stimulation strategies are different. This paper proposes various engaging strategies to optimize both LTE and STE of patients. Actually, proposes to combine the strategies concerning the whole rehabilitation program and the strategies to be applied during single rehabilitation exercises. Based on the proposed reasoning model, a cyber-physical computingbased solution for a personalized rehabilitation system is proposed. Various cyberphysical characteristics, such as function augmentation, adaptive and learning control, are being implemented in the system in order to realize the various strategies of engagement.