Engineers at the University of California San Diego recently developed a new way for robots to accurately navigate the indoors terrain, even without any recognizable feature and in poor lighting.
The technology includes sensors that use WiFi signals that help the robot map its direction and movement. It is a novel way to approach indoor robot navigation. In most cases, systems rely on optical light sensors like LiDARs and cameras. Here, the ‘WiFi sensors’ make use of radio frequency signals instead of visual cues, so they are still effective in conditions where optical light sensors would struggle— in poor or changing lighting and long corridors or warehouses.
The researchers noted that by using WiFi, the technology would be considerably cheaper than using expensive LiDARs.
This study, led by Professor Dinesh Bharadia, will be presented at the 2022 International Conference on Robotics and Automation (ICRA) which will occur in Philadelphia from May 23rd to 27th.
‘We are surrounded by wireless signals almost everywhere we go. The beauty of this work is that we can use these everyday signals to do indoor localization and mapping with robots,’ said Bharadia.
‘Using WiFi, we have built a new kind of sensing modality that fills in the gaps left behind by today’s light-based sensors, and it can enable robots to navigate in scenarios where they currently cannot’, added Aditya Arun, first author of the study.
The researchers built their model system making use of off-the-shelf hardware. The system is made up of a robot with WiFi sensors which are built from commercially available WiFi transceivers. With these devices, wireless signals can be transmitted and received from WiFi access points in the environment. The distinguishing feature of these WiFi sensors is that they use this constant communication with the WiFi access points to map out the robot’s location and direction of movement.
‘This two-way communication is already happening between mobile devices like your phone and WiFi access points all the time— it’s just not telling you where you are,’ said Roshan Ayyalasomayajula, PhD student and co-author on the study. ‘Our technology piggybacks on that communication to do localization and mapping in an unknown environment.’
This is how the process works. At first, the WiFi sensors do not know the robot’s location or where the WiFi access points are in the environment. As the robot moves, the sensors send out messages to the access points, using the replies as landmarks. Each incoming and outgoing wireless signal has its own special physical information— an angle of arrival and direct path to or from an access point— which can be used to determine the location of the robot and access points in relation to each other. The algorithms developed by Bharadia allow the WiFi sensors collect this information and make calculations. As the process goes on, the sensors gather more information and can accurately pinpoint the robot’s destination.
The technology was tested on a floor of the researchers’ office building. Several access points were placed around the space and affixed the robot with the WiFi sensors, along with a camera and a LiDAR for comparison. The robot travelled several times around the floor, going up and down long narrow corridors and passing dimly lit and bright spaces.
In these tests, the WiFi sensors were as accurate as the commercial camera and LiDAR sensors.
‘We can use WiFi signals, which are essentially free, to do robust and reliable sensing in visually challenging environments,’ said Arun. ‘WiFi sensing could potentially replace expensive LiDARs and complement other low cost sensors such as cameras in these scenarios.’
The researchers plan on combining WiFi sensors with cameras for a more complete but affordable mapping technology.
By Marvellous Iwendi.
Source: UC San Diego