If you are interested in the future of robotics and assistive technology, check out the following video and see our proposed approach in action!

Are you curious to see how a hybrid variation of the deep deterministic policy gradient (DDPG) reinforcement learning approach can reduce energy consumption while improving the performance of a 6 degree of freedom robotic arm in the target-reach task? Our recent publication proposed this approach, which employs a spiking neural network (SNN) for the actor model and a DNN for the critic one, and uses a combination of RGB and laser scan data for collision avoidance and object detection.


elEL
Μετάβαση στο περιεχόμενο