I’ve been interested in robotic exoskeletons for years, having done my PhD work in that area. More recently, I’ve become interested in the developing field of soft robotics and have been dreaming about use cases that combine elements of both.
Although professionally, I tend more to focus on B2B applications of technologies, as a consumer, I can dream… and as an avid visitor to the wilderness, right now I’m dreaming of “assisted” exploration of remote regions.
There are certainly existing exoskeletons on the market, whose function is to assist movement – either in a healthcare setting to, for example, assist with spinal cord injury rehabilitation, or in a factory setting to assist workers with repetitive overhead tasks that are hard on the body when maintained hour after hour, day after day. Eksobionics is one of the leaders in this area.
Some exoskeletons even harvest energy from the user’s motions: Together, the US military and several companies have been developing technologies in this area. The problem with these, for the use cases I’m interested in, is that they are rigid, bulky, somewhat difficult to control and use, and, in the case of those that don’t harvest energy but draw on energy provided by batteries, very heavy.
Looking at the next generation technology, what is particularly exciting to me is the prospect of harvesting energy from a soft wearable robot while engaged in exploratory activities, and then releasing that energy as needed.
Soft robotics is a relatively new subdomain of robotics. Unlike what most of us think of when we think of traditional robots (whether humanoid or not), soft robots are made of, well, soft materials. These may include soft but solid materials as well as fluids.
One of the reasons soft robots are of interest is to make them more accessible and perhaps less threatening than conventional robots made of rigid materials – more life-like and inspired by biology, and sharing more of the properties that natural organisms have.
The challenges in designing and building such robots are many, including control, power and materials. Controls is more my background, but for the purposes of dreaming about use cases, I’m currently more focused on power and materials:
What are the properties needed to make materials usable in soft robotics? For one thing, they need to be able to undergo fairly large elastic deformations. Imagine how much your skin stretches and deforms in the course of normal movements. Secondly, they need to be able to physically interact with their environment without being harmed. Again, think of your skin – although it’s delicate in many ways, it forms a remarkable barrier between (the rest of) you and the environment. In some ways, this is easier for a soft robot than a rigid one, but still, the outside surface needs to be durable enough to handle physical interactions. Of course, in the “natural” case, being self-healing helps too, but that might be trickier for a while yet, in a soft but artificially created robot.
What kinds of materials are most suitable for these requirements? They definitely include elastomers, which are polymers (natural or synthetic) exhibiting both viscosity and elasticity. Historically, these have been used, for example, as sealants or as somewhat flexible individual parts, in other (perhaps rigid) machinery. Combinations of materials are also used, such as elastomeric matrices with other materials embedded in them. I also just read about a very cool electric eel inspired, origami-folded hydrogel able to virtually instantly generate 100 volts of electricity (University of Fribourg, with a link to their recent Nature paper here). From a wearables point of view, this has incredible implications for the ultimate power source coming from within the human body, but not requiring the skeletal muscles. Hmmm, many of us carry around a little extra “stored energy”, whether we want to or not. I wouldn’t mind finding a way to put that to good use.
Soft wearable exoskeletons
In my use cases, harvesting and releasing energy is a critical functionality. One of the areas of active research in soft robotics is in materials science, specifically, the development of elastomers that can convert elastic deformations to electrical energy. Within soft robotics, these types of materials may be used as an artificial skin, covering other functional parts of the robot, which may be soft or rigid.
Alternatively, these elastomers may be used as artificial muscle, not as a skin covering – and may in fact constitute the whole robot, not just a muscle actuating other rigid parts. The development of artificial muscle materials is well established. The use in soft robotics – at scale – however, is newer and remains a key challenge.
For autonomous robots, there will be one set of functional requirements. CSAIL at MIT and other leading research labs have come up with amazing designs in this area. Here’s an incredible YouTube video of one of their recent robotic fishes, swimming around a reef.
To me, some of the most interesting applications are in cyber-physical-human systems, in this case, wearables that can “power-assist” my movement through the environment. Most likely, this means some combination of exo-skeleton (possibly including rigid materials), soft materials, and actuation provided by both the human wearer and the robot.
Interesting use cases
Much like electric bicycles are powered by a combination of human exertion and electricity, the wearable soft robot I’m envisioning won’t replace but will augment my own efforts. I’d like to see the energy provided by the wearable – at least in part – come from stored energy harvested previously in the course of my normal movements. In other words, harvesting energy that would otherwise have been wasted.
What could this look like? Imagine me backpacking or climbing all day in the wilderness. If I think of my robotic augmentation as a safety measure, perhaps I somewhere come across a situation in which I need a short burst of energy. Maybe I need to jump across a stream I couldn’t otherwise cross safely. Or maybe I need to haul myself up somewhere that I can reach but is outside of my zone of optimal strength. Alternatively, I might want to use my harvested energy as heat, not when I’m already sweating like crazy and over-heating from my exertions but at night or when the storm hits, and I’m freezing.
Maybe it’s not a safety measure at all, but maybe in each step I take my otherwise wasted energy is put to use in putting a spring in my step, enhancing speed or enabling me to take a slightly longer leap when I’m running. Here, the primary benefit to me may be more about covering more ground more rapidly.
Consequences and ethical considerations
I can imagine my less technologically oriented outdoor friends shuddering at the thought of yet more people using this technology to go further into the wilderness than they otherwise have the skills for, resulting in yet more people getting themselves into trouble. Yep, that’s going to happen.
It’s important in technology development to look at uses, even arguably bad consequences. Yes, people will do things that get them into trouble. However, this is really no more than what we’ve already observed: For example, people thinking that cell phones would enable them to go anywhere, because they can always (or so they think!) call for help and be rescued. This may be an unfortunate side effect for those of us who enjoy – and train in – the wilderness (and are involved in rescuing people who get into trouble there). But most people would not argue against the adoption of the underlying technology. In any event, search and rescue workers could use the same technology, to get to people in distress more quickly or more safely.
My wish list
To make it useful, my wish list for this technology includes:
- Enough energy harvested to provide a useful “boost”
- Comfortable to wear – it has to be integrated with the rest of my gear
- Washable – yes, that would be really nice if it’s going anywhere near my body as a wearable!
- Lightweight – otherwise, the extra effort required to carry it around will take away from its usefulness
- Length of time for energy storage – this will vary with the specific uses; if it’s primarily to give me a little bounce in each step, energy storage is not really a problem; if I want to save up that energy for a single-shot release of a larger amount later, perhaps hours or even a day or two that requires a separate solution
- Fully compatible with extremely active wear
In the longer term, the ultimate in coolness would be to integrate my soft exoskeleton with AR and with AI, but even well before that, there are fabulous use cases.
OK that’s fun dreaming for personal use. What about B2B applications for this? More to come on that topic.