Research project

Broad research project. Can integrate several other kind of projects and works

Bidep Locomotion in Sloped Environments - Combining Reinforcement Learning and Dynamical Systems

In this work, reinforcement learning techniques are implemented and compared to address biped locomotion optimization. Central Pattern Generators (CPGs) and Dynamic Movement Primitives (DMPs) were combined to easily produce complex trajectories for the joints of a simulated DARwIn-OP humanoid robot. 

People involved in this project: 
Project status: 
Past or closed project

Combining adaptive locomotion primitives

Adaptive locomotion primitives

Algorithms that made it possible for a robot to be capable of learning a movement that successfully accomplished the intended goals for a certain instance of a task were an important milestone in robotics and reinforcement learning. However, such capability alone is not enough in order to answer to most challenges when learning motor skills if the learned movement is not expandable to other instances of the task.

People involved in this project: 
Project status: 
Project in progress

Cycling Device

Development of a Cycling Device for rehabilitation of the lower extremities.
 
With population aging, there is higher susceptibility to strokesand neurodegenerative diseases. These conditions usually have sequels and, therefore, a necessity of rehabilitation. 

People involved in this research: 

A bio-inspired computational model of motion detection

This project is an innovative undertaking combining insights of computational neuroscience, neuroethology, computer vision and robotics. 

People involved in this project: 
Project status: 
Past or closed project

A bio-inspired computational model of motion detection

In many animal species it is essential to recognize approach predators from complex, dynamic visual scenes and timely initiate escape behaviors.

People involved in this research: 

Biped locomotion

The aim of this project is to develop a simple strategy for controlling and adapting bipedal locomotion using bio-inspired algorithms.

The problem of bipedal locomotion includes several challenges that must be adressed:

You are missing some Flash content that should appear here! Perhaps your browser cannot display it, or maybe it did not initialize correctly.

ASBGo walker

A mobility assistance Smart walker to infer the user's navigations intentions
Walkers play an important role, due to the large number of potential users, with a focus in the elderly, its simplicity and their ambulatory potential. They were designed to improve pathological gait, through a support base for the upper limbs that improves the balance of the individuals and reduces the load on their lower limbs. 

People involved in this research: 

Quadruped locomotion

The aim of this project is to develop a complete framework that can autonomously generate, adapt, control and plan complex motor behavior for articulated robots with many degrees-of-freedom (DOFs) .
We focus the topical issue of robust, flexible and adaptive goal-directed quadruped locomotion in unknown, irregular terrain, modeled as discrete, sensory-driven corrections of the locomotion rhythmic patterns.

People involved in this research: 
Syndicate content