There's now a self-driving car so 'micro', it's thinner than a human hair

The process of miniaturisation hits every technology sooner or later. From laptops, to smartphones, to smartwatches. And now to the self-driving car.

Engineers at the Harbin Institute of Technology in China and the University of California San Diego in the US have developed a "spherical micromotor" that can autonomously navigate through a maze. "We have embedded artificial intelligence into a micro/nanorobot," Longqiu Li, who led the project, told

"We introduce a smart microvehicle for precise autonomous navigation in complicated and dynamically changing environments through optimal path planning. Similar to their large vehicle counterparts, the autonomous navigation of microvehicles entails collision-free movement in dynamic environments."

All external

All that intelligence doesn't fit inside the robot itself, of course - it's all external. A camera scans the maze, and sends the data to an image processor which identifies obstacles and creates a map. 

Then that map is sent to an AI planner which determines the shortest collision-free path to the destination. Finally, the AI sends its route to a magnetic field generator, which actually steers the teensy car.

The actual routes taken by the tiny vehicles. Credit: Harbin Institute of Technology

The actual routes taken by the tiny vehicles. Credit: Harbin Institute of Technology

What it's for

Applications for the technology include the ability to deliver drugs, diagnose disorders or conduct precision surgery within the body. To do that, they'll develop features like autonomous braking, "cruise control", lane-keeping and interactivity with the environment.

"We want to apply the micro/nanorobotic systems in the field of biomedical operations and nanoscale manipulation," Li said.

The full details of the technology were published in ACS Nano.

Duncan Geere
Duncan Geere is TechRadar's science writer. Every day he finds the most interesting science news and explains why you should care. You can read more of his stories here, and you can find him on Twitter under the handle @duncangeere.