• One Thing Better
  • Posts
  • This Robot Dog Learned to Walk in One Hour. It Has Lots to Teach Us

This Robot Dog Learned to Walk in One Hour. It Has Lots to Teach Us

Researchers wanted to know: When animals learn to walk, what do they learn from stumbling?

This is a newsletter about how you can build a better future. Ready to seize tomorrow? Subscribe here — and don't forget to pre-order my book on how to be more adaptable in your career!

Researchers wanted to know: When animals learn to walk, what do they learn from stumbling?

To find out, they built a little robot dog, programmed it to learn from its mistakes, and then watched as it stumbled its way into gracefulness.

As I read about this research, I couldn’t help but think how relevant it is to us — which is to say, adults trying to learn and accomplish new and complex things. When we set out on a new adventure, we are often frustrated by our initial stumbles. Maybe we see them as failings.

But in fact, they are not failings at all. They are the lessons that enable our eventual success.

Before we get there, let’s walk the dog

We’ll start with the obvious: When a baby animal (and human) is born, it relies on spinal cord reflexes that are hard-wired into their bodies. But while those reflexes allow a baby to move, they do not give the baby much control.

That’s what grows over time. Complex movements must be practiced and repeated, as a network of neurons called a central pattern generator “produce periodic muscle contractions without input from the brain.” In other words, your body is learning to do tasks — walking, swimming, chewing — without you having to consciously think through them.

When a baby learns to walk, that central pattern generator is hard at work. Eventually it can guide a baby over a totally flat surface, but of course, the world is not a flat surface. That’s where things get interesting.

From the report:

A small bump on the ground, however, changes the walk. Reflexes kick in and adjust the movement patterns to keep the animal from falling. These momentary changes in the movement signals are reversible, or 'elastic', and the movement patterns return to their original configuration after the disturbance.

And the more that happens, the more a body learns how to adapt.

This is what researchers wanted to program into the dog: They wanted to watch how a robot reacted to bumps, and then, over time, learned how to adapt to them. So they built the little robot dog, which they named Morti, and gave it its own version of a central pattern generator (or CPG). Then off it went:

If the robot stumbles, the learning algorithm changes how far the legs swing back and forth, how fast the legs swing, and how long a leg is on the ground. The adjusted motion also affects how well the robot can utilize its compliant leg mechanics. During the learning process, the CPG sends adapted motor signals so that the robot henceforth stumbles less and optimizes its walking.

Robots learn faster than biological babies, which is why the dog learned to stride in about an hour. But of course, the robot was modeled after us. We are the original learners.

So, what can we learn from the dog?

When we try something new, we are bad at it.

That’s just a fact. It is unavoidable. We cannot be good at something at first, because we are a novice operating in a world of experts.

Therefore, our greatest challenge is not to be good. Our greatest challenge is our willingness to be bad.

This American Life creator Ira Glass spoke memorably about this, in an old interview with Current TV. Consider the way he puts it:

If you’re trying to do creative work, Glass says, it is because you have good taste. You know what good looks like and you want to do good work.

However, you have not yet developed the abilities to do good work.

This is the real problem: Your taste is great, your abilities are not great, and therefore you will create work that you know is bad. He says:

A lot of people never get past that phase. A lot of people at that point, they quit. And the thing I would just like say to you with all my heart is that most everybody I know who does interesting creative work, they went through a phase of years where they had really good taste and they could tell what they were making wasn’t as good as they wanted it to be — they knew it fell short, it didn’t have the special thing that we wanted it to have.

Ira’s advice: Just keep going. Get through that phase. It is normal and unavoidable. The most important thing, he says, is to simply “do a lot of work.” The more you do, the better you’ll get.

And eventually, your abilities will catch up with your taste.

This is the lesson of the walking robot dog. It was not given the gift of walking — instead, it was given the gift of learning. Every time it hit a bump, it fell and learned a little more about how to navigate bumps. It developed a sense for adaptability; it learned to adjust its plans, to catch itself mid-stumble, and then to revert back to a confident stride.

We cannot expect greatness in the beginning. Instead, we can expect bumps and falls.

But here’s the thing: Without those bumps and falls, we’d get nowhere.

Let's Connect!

Like what you read? Subscribe and take control of your future.

🕵🏻‍♀️ Looking for more lessons to help you change? Check archives.

💌 What do you think? Let me know!

👋 Say hello on Instagram, Facebook or Twitter.

📕 Pre-order my book to become more adaptable in your career!

Cover credit: Felix Ruppert / Dynamic Locomotion Group