Imagine the scene after a huge earthquake or natural catastrophe, such as the devastating events in Fukushima or Haiti. An injured victim is buried underneath the wreckage. After some jostling, a spotlight pierces the darkness, the sound of hydraulics and motors approaches, and the rubble is lifted safely clear by a rescuer who isn’t even human.
Advances in robotics are making many experts predict a near future where rescue robots will scour disaster zones en masse. But their success depends on the alignment of several disciplines.
First, a robot plunging into danger needs to power itself independently. These often very heavy devices require a lot of electrical power; the more power they have to carry on board, the heavier they are, which requires more power in turn, and so on. The solutions to this energy problem vary greatly. Boston Dynamics’ BigDog carries a one-cylinder, two stroke Go-Kart engine (like those used in lawnmowers), which drives 16 hydraulic motors in its legs. By contrast, NASA’s Opportunity rover can theoretically keep exploring Mars forever (provided the mechanisms still work) as it recharges itself with a solar panel.
Advances in robotics are making many experts predict a near future where rescue robots will scour disaster zones en masse.
In the world’s foremost robotics competitions, entrants can’t be tethered to external power or communications, and in the most rigorous tests, wireless communication is purposefully degraded to give them a chance to prove their self-help skills. While that seems tough, a city struck by a killer earthquake or a forest engulfed in flames will be a much greater challenge. Search and rescue bots will have to go deep into dangerous territory, cut off from human operators with patchy communication signals. It will be making its own decisions about what to do next, using machine learning and other AI algorithms to self-teach.
Pre-programming robots for unpredictable environments is incredibly difficult, but leaving a robot to its own devices would be dangerous. There’s a sweet spot to be found, and ‘learning to unlearn’ certain behaviours can be just as important in the field. Restrict self-learning too much and the simplest obstacle might become a fatal stumbling block, like a flight of stairs or a door handle. Trust a system too much to try new things and it might decide a disaster victim is another piece of rubble and cause more harm.
The other secret to a successful search and rescue operation is sensors, and there are as many kinds as there are environments they have to work in. With feedback from accelerometers or gyroscopes in multiple dimensions, motion sensors give the robot critical information like orientation to the ground – an essential input when scrambling over wreckage. It can also get information about its movements from load-bearing sensors, which measure shifts in weight. The motors – known as actuators – then compensate, moving the body in the opposite direction to keep it upright. For robots that are connected to operators at a home base, visual sensors are crucial too. Cameras – often two of them to provide a sense of depth – can show the operator what’s going on in the immediate area.
Although humanoid robots seem like a natural choice, the movements required for scrambling over wreckage are hugely complicated.
We can also design robots with sensors for dangers they’re likely to encounter in specific environments. Sandia National Laboratories’ Gemini-Scout is designed for mining accidents, finding and delivering provisions to survivors. As well as the ability to navigate rocky surfaces, debris, and even water and mud, it has a thermal imager to acquire video, a speaker and microphone for communication, and temperature and gas sensors so it can sense environmental hazards. Its devices are surrounded by explosion-proof casing, so if it’s surrounded by explosive substances, the robot’s electronics won’t spark to trigger blasts.
After the destruction caused by major disasters, getting from A to B to reach those in need can be difficult, demanding constant shifts in balance and weight that we humans do without thinking. Wheels are of limited use although unique configurations of movable wheel arrays are catching on. Designs inspired by quadruped animals like Boston Dynamics’ BigDog and Cheetah are also showing promise.
Although humanoid robots seem like a natural choice, the movements required for scrambling over wreckage are hugely complicated. Even standing upright is a demanding task for the robot’s processor and motors, as they try to imitate a human’s brain and muscles.
Disaster robots got their first real debut when they were sent into the incredibly difficult terrain of the World Trade Center towers following the September 11 attacks. They didn’t perform at all well, often getting stuck or breaking, but the test gave engineers a lot of real-world experience to work on the next generation of rescue bots.
However, after spending all that development time and money on a single machine only to have it crushed flat by a falling wall or run out of power at the worst possible moment and be lost forever, the answer might be to not put all your eggs in one basket. The solution for some environments might be an army of rescue robots, working as a team.
A group of robots has several advantages. If there’s a lot of thick concrete or metal at the disaster site, communication is likely to be very unreliable, so if the connection is lost with an individual bot, it can be maintained along a chain between those that are still in range. The command will be passed down the line to the unit at the front line. A swarm also allows for a distributed processing model. Each unit has their piece of the puzzle but is also aware of the outlook of every other bot and can take over the decision-making or operator-response should something happen to its nearby fellows. It’s a little like having one giant robot body and brain made up of small, fluid elements.
The solution for some environments might be an army of rescue robots, working as a team.
The members of a robot army don’t need to be identical. Several different kinds of bot can be deployed, each with its own talents. Larger, longer-range robots could carry smaller and more specialised devices like snakebots deep into a disaster zone to go to work.
One snakebot model, designed by Japanese robotics professor Satoshi Tadokoro, is nearly eight metres long and propels itself using nylon bristles powered by tiny individual motors. It only moves at a crawl of five centimetres per second, but it can climb 20-degree inclines, turn sharp corners and see what’s ahead with its front-mounted camera.
These low-powered snake-inspired robots are built for localised environments, but there’s a way round this. Potentially, longer-range models could carry them to the burned out factory or collapsed building and deploy them to map and report back on the environment.
Whatever the shape or size, the search and rescue robots of the future will accompany and assist humans in dangerous conditions, or may even be able to go it alone, leaving their human operators in the safety of the control room.
The shortfalls of robot rescuers
Humans can go all day on just a few meals but mechanical helpers don’t have anywhere near the energy efficiency or endurance of the human body. We also have the ability to adapt, which is what gives us such varied talents. Despite robots beating us in several criteria, such as tolerance for hazardous material, far-off vision and detailed spatial mapping, they tend to be over-optimised for one type of problem, and teaching them new things means expensive engineering and complicated programming.
Robots also don’t instinctively know how to be safe like we do, lacking situational awareness and context unless it’s programmed in advance. This is important in search and rescue scenarios where danger is ever-present. Future missions will likely see human and robot responders working together to augment each other’s talents.
This article first appeared in How It Works issue 82, written by Drew Turney.