Want to see some exciting biologically inspired robotic legs? Look no further than MIT's Robotic Leg Lab website. The site has several biologically inspired legged robot pictures and videos.
Kangaroos, Dinosaurs, Flamingos and Turkeys are all inspirations for legged bots. These mechanically correct, but cosmetically awkward, robots explore the ability for robots to jump, walk, and run. These three activates are actually quite different to control.
Stand up from your computer and take walk around the room. Notice that one of your feet is in contact with the floor all the time. Now try jogging. Notice that for short periods of time you are completely off the ground as you leap from one foot to the next. Now try jumping about. Your legs need to create upward momentum to leap and then they must absorb your weight as you land.
Jumping, walking and running are different because of what is called gate, or stride. Robotically controlling the different gates is difficult because it requires compensating for weight and balance. As human we do this quit easily…well for adults that is.
Another fun aspect of the MIT leg lab site are the simulations. The Cockroach Sim shows cockroaches scurrying around the floor. It’s life like, so try not to shiver too much. The “On The Run” video is comical as two robots run with kangaroos!
Have you noticed that there are several reassembling robots in the news and on the web? Discover Magazine highlights ckBot, a Rubiks Cube looking robot from the U Penn, that can reconfigure after being kicked apart. A self healing chair on U-tube a chair can break apart and reassembly itself. These may seem like parlor tricks, but they are really laying a foundation for a larger problem related to space exploration.
The key to making space exploration more affordable is reducing the cost of getting objects into space. Many scientists are testing the theory of launching several small objects and then assembling them in earth’s orbit to create a larger vehicle, station, or satellite. This is the premise behind the International Space Station, which is assembled by Astronauts. To make this theory cost effective, the vehicles must be assembled autonomously, which is why these small reassembling robots are so important. These projects can lead to great understanding for docking robots in space.
The Orbital Express program highlights the difficultly of assembling vehicles while orbiting the earth and without human interaction. This government funded program made a failed attempt at docking two satellites with a robotic arm. A failure in the navigation system caused the two robots to hit each other and bounce away.
There are several reasons for this project's failure that are common to all space operations. The remote distance of scientist and hardware creates problems. Also, the space environment itself creates challenges. Space has low gravity which makes holding two objects close together difficult because they tend to float away. Also, because of no atmosphere, surfaces that face the sun become extremely hot and surfaces in shadow are extremely cold. This can make material warp and can make connecting to interfaces difficult or impossible.
Add these space factors to the challenges of coordinating a group of robots and the task becomes even more difficult. Controlling a group of robots to work together autonomously is called swarm robotics. Swarm robotics has challenges of its own, which include getting robots to communicate with each other, knowing where one robots is relative to another, and determining mating locations. These reassembling robots are really solving the problem of swarm robots.
The information learned from these reassembling robots can be help space application. Think of a reassembly robot that can operate in a desert location. How different is that from operating on the moon? Two flying robots that become one is a very similar operation as docking two satellites. These reassembly robots are giving scientists better understanding of docking robot swarms, without the expense of sensing things into space. So next time you laugh at these parlor tricks, think again!
Science researchers are studying the way fiddler crabs migrate from nest to sea. Researchers replaced the sandy terrain with a slippery material and found that the fiddler crab was coming up short from the nest. It was found that the fiddler crabs move from one location to the next not by seeing or sensing their surroundings, but by calculating how much energy their muscles and body were used for locomotion. This would be an excellent method for small UAVs to navigate rocky and uneven terrain for two reasons: tire slippage and energy conservation.
UAVs are often used in sandy or rocky terrain where it is easy for tires to slip. This is especially problematic because most wheeled robots use high tech odometers to collect distance information from the number of wheel revolutions. A slipping wheel distorts the distance that the robot has traveled. Like the crab, UAVs that have slipping tires will not arrive at their destination. Using the Fiddler crab’s navigation method, travel distance can be better predicted and therefore make for a better performing UAV.
The crab scientists found that crabs optimize for energy because the speed of the crab can vary. The crabs are capable of moving quickly or slowly, and therefore adjusting their stride lengths. This means the do not simply count steps. Researchers believe crabs monitor the amount of energy they expel over the duration of the journey.
Knowing a UAV’s energy usage on long endurance trip is very important because many UAVs use batteries. UAV’s must constantly conserve energy. Energy level can be important to decide what tasks can be completed or what speed to complete them.
Currently it’s not ground robots that are monitoring power levels, it’s air vehicles. High altitude solar powered aircraft such as Helios and Centurion move to different configurations to collect more sun rays. Because their main power source is the sun, they must constantly monitor usage. If an aircraft looses power, it doesn’t simply get stuck in the sand. It falls out of the sky. Using this type of method in ground robots may be beneficial for long trips.
In order to navigate using energy usage, more sensors and computers are required to be on the robot. This can add more complexity to the robot rather than simply counting wheel revolutions. However, for trips over varied terrain where precise locations are required, this may be the best method. So the best way for your UAV to travel may be the crab walk.
"We live in a society exquisitely dependent on science and technology, in which hardly anyone knows anything about science and technology." - Carl Sagan
I've worked as a research scientist for the last 6 years at several governmental labs. I have a Masters in controls and specialize in robotic design. Biologically inspired robots and mechanisms have been a passion of mine since university, where I designed a grasshopper inspired robotic hopper. Since then, I have worked on space related mechanisms including wing warping and adaptive control. I am currently writing about science and working on new research projects in the Boston area.
Posts
