Engineers create low-cost four-legged software robots

Engineers at the University of California, San Diego have created a lightweight, low-cost four-legged soft robot that can maintain system control and movement through a pneumatic circuit and does not require any electronic equipment to drive. The project was led by Professor Michael T. Tolley of the University of California, San Diego, and was published on the cover of Science Robotics.

The reason why software robots have received a lot of attention is that they can operate safely in the vicinity of people and have a strong ability to adapt to the environment. Dr. Dylan Drotman from the University of California, San Diego said: "This work represents an important step towards a fully autonomous, electronic-free walking robot."

Software robots are constantly upgrading and iterating, from only running to having an inflatable gripper, which is capable of handling various items. Inflatable components and air pressure play a vital role in it. Most soft robots are powered by compressed air and controlled by electronic devices. However, this method requires components outside the robot's body, such as circuits, valves, and pumps. Generally, the components that make up the robot's brain and nervous system are bulky and expensive.

The software robot of this project is different. Its pneumatic circuit system control is composed of pipes and soft valves. It can judge according to instructions or signals sensed from the environment, and then walk. The robot is driven by the spine, and its computing power is similar to that of mammals. The research team was inspired by neural circuits found in animals, which control walking and running actions by generating rhythms.

The research team added components needed to realize advanced functions (such as walking) on ​​the basis of the oscillator and sensor of the pneumatic valve. Pneumatic valves have the advantages of low price and lightweight, and are easier to manufacture than electromechanical valves, and are usually used to control software robots. With proper design, these components can create complex fluid combinations that can control the circuitry of the legged robot's gait without any electronic equipment.

This robot consists of a system of four cylindrical chambers. To simulate the function of a generator, engineers built a valve system that can act as an oscillator to control the order in which pressurized air enters the robot's limbs by pneumatic muscles. According to the team, using this method can create a very complex robot brain, and their focus is to create the simplest pneumatic nervous system needed to control walking.

It is also reported that the robot's gait is inspired by the side-necked turtle. For this reason, the researchers built an innovative component that coordinates the robot's gait by delaying the injection of air into the robot's legs. When the valve is opened and closed in a specific order, it can be used. Realize walking. At the same time, there is a micro-bubble sensor device at the end of the arm frame of the soft robot body. That is to say, when the soft robot encounters an obstacle, the bubbles in its device will appear concave, prompting the valve to flip and move in the opposite direction.

In order to test the control of the omnidirectional motion of the soft robot, the research team developed a controller to operate two 4/2 valves. By changing the corresponding valve state, the robot is commanded to move forward, left, backward, and then right.

Each leg of the soft robot can be bent at three angles of motion. The legs are tilted down 45 degrees and are composed of three parallel and interconnected pneumatic air chambers with bellows. When the chamber is pressurized, the limb will bend in the opposite direction.

The researchers paired the diagonals of each leg with each other to simplify control. A soft valve switches the rotation direction of the limbs between counterclockwise and clockwise. The soft valve is equivalent to a double-throw switch, that is, a switch with two inputs and four outputs, so each input has two corresponding to it. Output. This mechanism is a bit like requiring two nerves and exchanging their connections in the brain.

In a nutshell, this research is a breakthrough in low-cost robotics technology, which can be used in areas where electronic equipment such as mines and nuclear magnetic resonance imaging machines cannot be used. The research team expressed the hope that on the basis of this research, the robot's gait can be improved so that it can walk freely on more rugged ground.