Beep boop - this is a robot. A new show has been posted to TWiT…
What are your thoughts about today’s show? We’d love to hear from you!
I watched Monday’s OfficeHours.global #1602 video about physical factors of human robots mentioned today by Alex. In Office Hours @39:29 Alex notes the tremendous adaptability of humans – our ability to deal with a vast variety of terrain. LLMs are interesting to me, but my passions and curiosity are ignited by that adaptability – our biomechanics.
One facet of our biomechanics not discussed was the hardness/softness of robots: how human robots can be rigid enough to routinely lift/manipulate up to ~100kg of goods while being soft enough to safely interact with humans. MIT Professor Neville Hogan has been researching this through his entire career. One example is Adaptive Control of Mechanical Impedance by Coactivation of Antagonist Muscles (1984). Hogan notes that co-activation of agonist and antagonist muscles in mammals is used to achieve rigidity, fluidity, and everything in-between. The flexors/extensors of the lower arm are an example. When all those flexors/extensors are activated simultaneously, we have a [rigid] fist. When those muscles are simultaneously relaxed, the hand softens and has fluidity/play. We have the rigidity to exert forces to break boards; we also have the softness to touch/hold/support a baby. Our robots must develop this capability! We as humans should constantly develop our capacity to be rigid/fluid. If anyone has questions how/what I do, feel free to ask…
The concept and mathematics of mechanical impedance is borrowed from Oliver Heaviside’s equations of electrical impedance. The concept is the same; the term impedance analogy is used to note the broad applicability of the idea to non-electrical systems. The crucial factor in any system to apply the impedance model is stored energy.
To me, the most interesting thing about Professor Hogan’s research is not the application to robots. The interesting thing is how rarely mechanical impedance – musculoskeletal impedance – is used to model/visualize our posture, movement, and ability to do work. Hogan used biomimetics to create his “soft robots”, but anatomists rarely use impedance to describe the rigid/soft nature of our structure. Most of us will never design a robot, but each of us has to deal with managing the function/development/happiness of our bodies. Isometric contraction is an obscure term, but co-activation (of opposing muscule pairs) is an accessible term to visualize. Why are health/fitness professionals only taught the i-word? How does that ever explain anything? Co-activation explains so much!
Modeling biomechanics without including stored energy in the system is problematic. Our IT band works by storing/releasing spring energy. Note the wonderful gliphy from that article:
See how the recovery happens through the spring energy being released. That brilliant accompanying science paper was written by evolutionary biologists and not by mainstream biologists/anatomists. In order to progress the knowledge, we had to do an end-run around the “keepers of knowledge” in this field.
I occasionally follow Office Hours Global, but I hadn’t heard about this particular session until Alex mentioned it on MBW. I’ll note my comments to Alex and suggest the organization invite Neville Hogan to participate in a session of Office Hours.
Good show. fun and informative. @Leo the “attic studio” looks great.
2 Likes