Tomislav Bacek

Learn More
Forecasting performances of feed-forward and recurrent neural networks (NN) trained with different learning algorithms are analyzed and compared using the Mackey-Glass nonlinear chaotic time series. This system is a known benchmark test whose elements are hard to predict. Multi-layer Perceptron NN was chosen as a feed-forward neural network because it is(More)
A novel modular variable stiffness actuator (VSA), for use in the knee joint of lower limb exoskeletons, is presented. The actuator consists of a combination of a spindle-driven MACCEPA (Mechanically Adjustable Compliance and Controllable Equilibrium Position Actuator) and a spring acting in parallel, (dis)engaged by means of a simple on/off mechanism(More)
The ankle joint is the main contributor in providing support to the body, ensuring forward progression and initiating the swing of the leg during the push-off phase of walking. However, its capability can be negatively altered by neuromuscular disorders. In these cases, an active ankle-foot orthosis (AAFO) can greatly enhance the rehabilitation of the(More)
In this paper, an improved Levenberg-Marquardt-based feedforward neural network, with variable weight decay, is suggested. Furthermore, parallel implementation of the network on graphics processing unit is presented. Parallelization of the network is achieved on two different levels. First level of parallelism is data set level, where parallelization is(More)
This paper presents design of a novel modular lower-limb gait exoskeleton built within the FP7 BioMot project. Exoskeleton employs a variable stiffness actuator in all 6 joints, a directional-flexibility structure and a novel physical humanrobot interfacing, which allows it to deliver the required output while minimally constraining user's gait by providing(More)
  • 1