Walking ain’t easy
Robotics engineers trying to copy human bipedal locomotion have a long way to go
For nearly 50 years, engineers worldwide have been trying to build humanoid robots capable of stable and efficient walking as humans do. While creditable progress has been made, they are still far from achieving that goal.
Even just to satisfactorily emulate only one of the body’s joint-and-muscle systems used in walking—i.e. the ankle/foot complex, or the knee, or the hip/pelvis complex—presents huge engineering challenges.
An ideal robotic foot for example, as a recent review paper pointed out,1 would have at least these three human characteristics: a flexible heel (shock absorption), toe flexion (energy efficiency, staying upright when standing still), and a flexible arch capable of being stiffened at critical points during each stride (energy harvesting—mimicking the windlass mechanism of the human foot—and shock absorption).
Meanwhile a robotic ankle needs to be able to store, and release, energy at appropriate times during each stride, to attain the energy efficiency of the human ankle.
What’s the progress so far towards implementing these features of the ankle/foot complex? “Due to their technical complexity,” the review paper laments, “all these human-like properties are still not available in one single ankle–foot solution.”1
Furthermore, the overarching challenge facing engineers is the interplay between the various components involved in human bipedal locomotion—a key to stability and efficiency. E.g., the knee, site of action for many of the leg muscles crucial to forward propulsion, plays a central role for the energy transfer between knee and hip and ankle. And in robots it has been demonstrated that actuation at toe-joint can usefully lower the speed at which knee joints have to move, and increase walking speed and step length.
But the interplay is even more complex2—not just between mechanical and morphological (structural) mechanisms, but neural (nerve/signalling) systems as well. The review paper says of this:
“As a result, humans show very robust, versatile and energy efficient functional abilities in a vast range of locomotion conditions. The process of transferring such principles into robotic platforms is not trivial, since the complex interplay between the sensorimotor mechanisms involved in human walking is still far from being fully understood.”1
“Far from being fully understood.” No wonder bipedal prototypes so far cannot cope with uneven/unstructured terrain, and have variously suffered from “unnatural motions”, “high-energy costs”, “high computational demands”, “rigidity”, or being “rather sensitive to external perturbations”. So no wonder that still, “a human is the golden standard for bipedal walking efficiency.”1
Consider, then, the source of that ‘golden standard’. Think of the goal-oriented and intensive thinking of keen engineering minds trying at length to copy that standard. Surely this speaks against the notion that some goal-less, thought-less, mind-less, ‘natural’ process can be credited with having changed some supposed quadrupedal (four-footed) ancestor into us. Rather, as the Bible says, God created us with the ability to walk—ideally, with Him (e.g. Genesis 5:22, 24; 6:9; 2 Corinthians 6:16).
References and notes
- Torricelli, D. and 9 others, Human-like compliant locomotion: state of the art of robotics implementation systems, Bioinspir. Biomim. 11(5):1002, 22 August 2016. Return to text.
- Even arm-swinging is integral to the energy efficiency of the human gait. See: Arm-swinging not an “evolutionary relic”, Creation 32(2):8, 2010; creation.com/arm-swinging. Return to text.