Innovation That Matters

Mars rover | Photo source WikiImages from Pixabay

Tech Explained: Autonomous Robots

Tech Explained

How autonomous robots are able to work independently

Autonomy is the ability to make independent decisions. Autonomous robots, like humans, can make their own decisions and act on them. One of the most common autonomous robots is the Roomba vacuum cleaning robot. Like other autonomous robots, the Roomba uses sensors to perceive its environment and takes appropriate actions based on these perceptions. It can be left alone to do its job without any help or supervision from humans.

So, how do autonomous robots work and how else are they being used?

How autonomous robots work: As with the Roomba, most autonomous robots have tools that allow them to perceive their environment, make decisions and carry out decisions (also referred to as actuation).

Autonomous robots sense their environment through a variety of technologies including laser scanners, cameras, microphones, force-torque sensors and spectrometers. Simpler autonomous robots such as the Roomba rely on infrared or ultrasound sensors to help the robot “see” obstacles in their path. Higher-level robots such as autonomous vehicles use more complex sensors like cameras, radar and lidar (a detection system like radar, but using light from a laser). Combined with image-recognition software, these sensors allow the robot to precisely identify and categorise the objects they “see”, and make real-time “decisions”.

Some autonomous robots are designed to work in a constrained environment. For example, lawn-mowing robots might use buried border markers to define the limits of a yard so they don’t mow the entire neighbourhood (or run over the neighbours’ cat). A cleaning robot might use a GPS to map a building and manouever from point to point. Robots designed to explore other worlds might use sensors to “build” a map of the area as they travel.

Autonomous robots have on-board computers, but they may also connect to the Internet to download information and upload updates. “Self-learning” robots, also called adaptive or intelligent robots, use on-board artificial intelligence software to learn from their environment and adapt their behaviour. One example of this is Aibo, a pet robot from Japan that adapts to its environment. Aibo can learn to shake, for example.

Benefits: Autonomous robots are already being used for a wide variety of tasks. Their use is growing daily as more sophisticated robots are developed. Uses range from cleaning dangerous places, such as sewers and nuclear plants, to delivering lab results and patient samples in busy hospitals. Autonomous robots are also being used to conduct research. For example, researchers at Virginia Tech created a five-foot-long autonomous robotic jellyfish designed to monitor the underwater environment.

Things to keep in mind: At the moment, most autonomous robots need some human assistance to help them navigate their environment. For example, self-driving vehicles are not yet at the point where they can operate without a human “backup”. But as the technology grows more “intelligent”, this is likely to change. Soon, autonomous robots will be able to carry out many tasks that today require humans such as making deliveries, farming, research, construction and even some forms of law-enforcement.