Research in the field of humanoid robotics is of great importance as it can potentially be used in the development of prosthetics, exoskeletons, and rehabilitation devices. An important research problem in generating stable bipedal gait is the balance maintenance during walking or the ability to prevent the humanoid robot from falling. Prevention of falling is of great importance in the practical application of humanoid robots.
In this project, NeuraBASE learns the sensor events obtained via the force sensors and, the accelerometer used to control the motor events of the Bioloid’s Dynamixel motors.
A controller network is subsequently constructed to store the associations between the sensor neurons and the motor actions. The strengths of these associations are adjusted through reinforcement learning to enable NeuraBASE to learn the most appropriate motor actions for a given sequence of sensor events.