How do robot vacuums navigate around your house?
In recent years, robot vacuums have become increasingly popular as a convenient and efficient way to clean floors without the need for manual labor. These intelligent machines use a variety of sensors and algorithms to navigate their way around a room, avoiding obstacles and cleaning up dirt and debris.
But how do they do it?
In this article, we will take a closer look at the technology behind robot vacuums and explain how they navigate to keep your floors clean. Whether you’re a robot vacuum owner or simply curious about the latest in home cleaning technology, this article will provide a comprehensive overview of the fascinating world of robot vacuum navigation.
How Do Robot Vacuums Navigate?
When robot vacuums first came out, about 15 years ago, they moved around at random. This meant that they cleaned some areas twice and missed other areas altogether.
But the technology has come a long way since then. Today’s robot vacuums clean your house systematically, following lines like you do when you mow a lawn.
And if they run out of charge halfway through a cleaning job, they return to base, charge up, and then go back to where they left off.
Robot vacuums find their way around your house using two different processes: mapping and navigation.
You can think of mapping as “where do I need to go?”, and navigation as “how do I get there?”
Let’s look more closely at these two key aspects of how robot vacuums navigate.
Robot Vacuum Mapping
Before a robot vacuum can navigate around your house, it first needs to create a map of your house. These maps allow the robot vacuum to ensure that it cleans every part of a room and every part of a house.
These maps also maximize the robot vacuum’s efficiency by ensuring that it completes a job in the minimum necessary maneuvers. And that in turn extends battery life and means they can clean more area between charges.
Robot vacuums use one of two different technologies to map a house: LIDAR or VSLAM.
Mapping with LIDAR
LIDAR stands for “light detection and ranging” and is the most precise of the two technologies. LIDAR uses an invisible spinning laser beam that emits waves of pulsed light in the surrounding area.
These pulses of light bounce off surrounding objects and return to the sensor. The sensor measures the time it took for each pulse to get back to the sensor to calculate the distance traveled. And it does this millions of times per second. It’s similar to how radar works.
An onboard computer then converts those millions of data points into a 3D map which the robot vacuum uses to navigate around your house.
Mapping with VSLAM
VSLAM is an acronym for “visual simultaneous localization and mapping”. Whereas LIDAR uses laser beams, VSLAM uses camera images.
At the core of VSLAM is an algorithm called SLAM (simultaneous localization and mapping).
This algorithm allows a robot to map out unknown environments. It does this by using the information it gets from cameras and/or from the number of wheel revolutions. VSLAM robots have light sensors that measure wheel revolutions and they use this information to determine how far they have traveled in a particular direction.
VSLAM is not as accurate as LIDAR but it is more affordable as it is based on cameras rather than infrared beams.
The map that robot vacuums create, using either LIDAR or VSLAM, is used by the robot vac to clean more efficiently: it can work out the paths it needs to follow in order to cover a whole room.
The first generation of robot vacuums didn’t have mapping. This meant that they would move around a room at random: they would clean some parts of a room twice and miss other parts altogether.
Most robot vacuums today have either LIDAR or VSLAM which ensures they move around your house in a deliberate fashion, cleaning everywhere in a room and never cleaning the same spot twice.
Five Key Aspects of Robot Vacuum Navigation
Robot vacuum navigation consists of a handful of behaviors that are triggered by sensors in the body of the robot vac.
1. Robot Behaviors
Looking at a robot vacuum as it navigates around your house, it can seem like a marvel of artificial intelligence.
But robot vacuums are actually modelled on the behaviors of ants which.
Researchers at MIT realized that while ants don’t have enough enormous brains, they are able to successfully avoid obstacles using just four behaviors:
- “straight” behavior tells the robot to keep going straight until it hits something
- “bounce” behavior tells it that when it hits something, it should stop, turn to an angle away from the wall and move straight again
- “spiral” behavior tells it to move outward in a spiral, cleaning the floor in expanding circles
- “wall-following” behavior tells the robot vacuum to follow the wall by “bouncing” and going “straight” until the wall is constantly a certain distance away
Of course, these four behaviors require sensors, which are the mechanisms that allow robot vacuums to navigate.
2. Bump Sensors
Bump sensors are located on the leading edge of the robot vac. When the vacuum bumps into a chair or a table or some other object, these sensors tell the vacuum to stop, turn at a certain angle, and move in a straight line again.
3. Cliff Sensors
Cliff sensors measure the distance between the base of the robot and the floor. They do this by sending an infrared beam to the floor and then measuring the time it takes to bounce back. If the beam takes longer than normal, the robot vacuum knows it has found a cliff and needs to retreat.
4. Wall Sensors
Wall sensors work like cliff sensors: they send an infrared beam sideways, to measure the distance between the side of the robot and the walls. They help the robot vacuum to follow the line of a wall.
5. RCON Sensor
In addition to these three types of sensors, Roomba vacuums also have a ‘Room Confinement Sensor’. This sensor looks like a mini-tower and is located on the top of the Roomba.
The Room Confinement Sensor (RCON) allows the Roomba® to detect compatible accessories, such as the Home Base® and Virtual Walls®.
How do robot vacuums navigate?
They use a combination of mapping and sensors to navigate their way around your house.
Robot vacuums build virtual maps that tell them where they need to go. And they use sensors and a small repertoire of behaviors to avoid obstacles and navigate their way around your house, based on the map they have built.
Unlike the robot vacuums of 15 or 20 years ago, today’s robot vacuums clean your house systematically, making sure they don’t cover the same area twice or leave any areas uncleaned.