Works on the scale of insects can squeeze into places where their larger counterparts cannot, for example deeply in a collapsed building to look for survivors after the earthquake.
However, when they move through debris, compact crawling robots may encounter high obstacles that they cannot climb or sloping surfaces that slide. While aviation works could avoid these threats, the amount of energy required for flight would seriously limit how the robot can travel to the wreck before he had to return to the base and charged.
To get the best of both transport methods, MIT scientists have developed a jumping robot that can jump on high obstacles and jump over sloping or uneven surfaces, using much less energy than an air robot.
A jumping robot, which is smaller than human thumb and weighs less than a paper plot, has a elastic leg that drives it from the ground, and four fluttering modules that give him lifting and control its orientation.
The robot can jump about 20 centimeters into the air, i.e. its height four times, at a side speed of about 30 centimeters per second, and there are no problems with ice jumping, saturated surfaces and uneven soil, and even on a drone floating. The hop robot consumes about 60 percent less energy at all times than his flying cousin.
Due to the slight weight and durability as well as the energy efficiency of the jumping process, the robot can carry about 10 times more load as a similar size of a slope, opening the door to many up-to-date applications.
“The possibility of placing on board batteries, circuits and sensors has become much more feasible in the case of a jumping job than flying. We hope that one day this robot can leave the laboratory and be useful in scenarios in the real world,” says Yi-Hsuan (Nemo) Hsiao, student of myth and co-author of paper in robots.
Hsiao joins the article by the authors of Sondnan Bai, scientific assistant professor at the University of Hong Kong; and Zhongtao Guan, the upcoming student of myth who completed this work as a visiting student; and also Suhan Kim and Zhijian Ren from MIT; and older authors of PakPong chirattananon, associate professor at the University of City in Hong Kong; and Kevin Chen, associate professor at the Department of Electrical and Computer Science and the head of the Tender and Micro Robotics laboratory at the Electronics Research Laboratory. Test appears today in
Maximizing performance
Jumping is common among insects, from fleas, which will jump on up-to-date hosts for field horses, which are associated around the meadow. While the jump is less common among insect robots that usually fly or crawl, jumping provides many advantages of energy efficiency.
When the robot jumps, it transforms potential energy, which comes from its height outside the earth, into kinetic energy as it falls. This kinetic energy transforms back into potential energy when it hits the ground and then back to kinetics when it rises and so on.
To maximize the performance of this process, the MIT robot is equipped with a malleable leg made of compression spring, which is similar to the spring on the pen. This spring, it transforms the speed of the robot down into speed when it hits the ground.
“If you have a perfect spring, your robot can simply jump without losing energy. But because our spring is not perfect, we use flutter modules to compensate for a small amount of energy that it loses when it contacts the ground,” explains Hsiao.
When the robot reflects in the air, the flutter wings provide an elevator, while ensuring that the robot remains straight and has the correct orientation for the next jump. Its four fluttering mechanisms are driven by supple cylinders or artificial muscles that are robust enough to endure repeated impacts with the ground without damage.
The key to the robot’s performance is the quick control mechanism, which determines how the robot should be oriented on the next jump. Sensing is performed using an external motion tracking system, and the observer algorithm calculates the necessary control information using sensor measurements.
When the robot hopes, he follows the ballistic trajectory, cut in the air. At the top of this trajectory, he estimates his landing position. Then, based on your target landing point, the controller calculates the desired start speed at the next jump. While in the air the robot ends the wings to adjust its orientation so that it hit the ground at the right angle and wasp to move in the right direction and at the right speed.
Durability and flexibility
Scientists set up a jumping robot and its control mechanism to test on various surfaces, including grass, ice, saturated glass and uneven soil – he successfully traversed all surfaces. The robot could even jump onto the surface that tilts dynamically.
“The robot does not really care about the angle of the surface on which it lands. As long as it does not slip when he hits the ground, it will be good,” says Hsiao.
Because the controller can operate many areas, the robot can easily go from one surface to another without losing the rhythm.
For example, jumping through the grass requires more pushing than jumping through the glass, because the grass blades cause a damping effect, which reduces its jump. The controller can pump more energy into the robot wings during the air phase to compensate.
Due to its compact size and lightweight weight, the robot has an even smaller moment of inertia, which makes it more agile than a larger robot and better withstand collisions.
Researchers showed their agility, showing acrobatic reverse. A featherweight robot can also jump on the drone in the air without damage to any device, which can be useful in cooperation.
In addition, while the team demonstrated the bouncing robot, which wore twice the weight, the maximum load can be much higher. Adding more weight does not harm the robot performance. Rather, the spring performance is the most critical factor that limits how much work it can wear.
Moving forward, scientists plan to operate their ability to wear weighty loads by installing batteries, sensors and other robot circuits, in the hope that it will allow him to autonomously jump outside the laboratory.
“Multimodal robots (those connecting many traffic strategies) are generally difficult and particularly impressive on such a small scale. The versatility of this small multimodal robot-transmission, jumping over the rough or moving area, and even another robot-so he is even more impressive,” says Justin Yim, assistant professor at the university Illino He was associated with this work. “Continuous jumping shown in these studies enables agile and efficient transport in environments with many large obstacles.”
These studies are partly financed by the American National Science Foundation and the Mit Misti program. Chirarattananon was supported by the Council of Research Subsidies of the Special Administrative Region in Hong Kong, China. Hsiao is supported by the MathWorks scholarship, and who is supported by the barking scholarship.