Researchers at King's College London in United Kingdom have developed for the first time a way to send complex instructions to a robot without electricity, which can free up more "thinking" space for the robot's "brain". This world-first opens up the possibility of a new generation of robots. The findings were published in the latest issue of the journal Advanced Science.
Mimicking how certain parts of the human body work, the researchers used a new compact circuit that transmitted a series of instructions to the device through changes in the fluid pressure inside. By offloading the work of the software to the hardware, the new circuitry frees up computational space for the robot to "think", and the space that was originally used in the control center can be used to run more complex AI software.
In simple terms, the robot is divided into two parts, the brain and the body, according to the researchers. AI brains can help manage a city's transportation system, but why can't many robots even open a door? The reason for this is that the hardware has not kept up with the rapid development of software. By creating a hardware system that runs independently of the software, a lot of computing tasks can be offloaded to the hardware, just as the human brain doesn't need to tell the heart to beat.
Currently, all robots rely on electricity and computer chips to run. The "brain" of a robot is made up of algorithms and software that relay information to the body or hardware through an encoder and then perform the action. In the field of soft robotics, this problem is particularly prominent. The field uses soft materials to make devices such as robotic muscles, often introducing hard electronic encoders and putting pressure on the software to make the materials behave in complex ways.
The researchers have developed a reconfigurable circuit with an adjustable valve. It is placed in the robot's hardware, and the valve acts like a transistor in a normal circuit. Engineers can use pressure to send signals directly to the hardware, mimicking binary code, allowing the robot to perform complex actions without the need for electricity or instructions from a central brain. This allows for a higher level of control compared to current fluid-based circuits.
This result is expected to lead to the development of robots that can operate in environments where electrically driven devices cannot work, such as surveying in radiation zones or working in power-sensitive environments such as magnetic resonance imaging rooms.