Neuro-Robotics外文电子书籍.pptVIP

Coming Up April 22: Brain Control and Augmentation The Dark Side of the Force April 28: Google, Microsoft, and the Future of Medical Informatics Who is reading your medical records? April 30: Final assignment presentations Bibliography CONTROL OF POWERED UPPER LIMB PROSTHESES. P. A. Parker, K. B. Englehart, and B. S. Hudgins. Brain–machine interfaces: past, present and future. Mikhail A. Lebedev and Miguel A.L. Nicolelis. Prosthesis, Exoskeletons, and Rehabilitation: Now and for the Future. B. Dellon and Y. Matsuoka. Neuro-Robotics Machines of Loving Grace Spring 2008 Neuro-Robotics Blending of research in robotics, neurology, and artificial neural networks Motivations Model robot motor systems on the methods used in humans and animals: enable fast fluid movement Contrast static dynamic stability in humans and today’s robots Enable direct neural control of robotic limbs and bodies In order to interpret neural signals, we need to be able to model how they work Topics Robotic exoskeletons Brain control of robotics limbs Robots that learn to walk and grasp Controlling animal brains with neural implants Augmenting human and animal brains with artificial neurons Robotic Exoskeletons Enhance strength of user Compensate for muscular degenerative disease Provide superhuman capabilities Early work 1965: General Electric Research Development: Hardiman 1 Weighed 1500 lbs Could lift 750 lbs Attempting to operate both legs at once leads to “violent and uncontrollable motion” Making Exoskeletons Practical Problem was not just size weight: key problem was control Hardiman controlled by operator’s arms, hands, and legs pushing against switches The delay between the operators intent and the exoskeleton’s motion made dynamic balance difficult or impossible Intent to move Muscles start contracting Limb moves Switch Closes Robot moves EMG Control Can detect early activation of muscle using electromyograph Modulate robot motion to move at precisely the same rate the real limb mo