WiFi System for Robot Navigation

History

2022: System Announcement

At the end of May 2022, engineers at the University of California, San Diego, also developed low-power technology that helps robots map the environment indoors. The system helps robots navigate even when light is low or there are no recognizable landmarks.

The team of researchers belongs to the Wireless Communication Sensing and Networking Group, which is led by Dinesh Bharadia, professor of electrical and computer engineering at the University of California, San Diego. The developed technology has sensors that rely on Wi-Fi signals, which allows the robot to determine its environment and path. The system is a completely different approach to the navigation of robots in the premises, the technology is unique compared to previous ones using optical light sensors, such as cameras and lidars.

A system has been created to help robots "navigate" Wi-Fi buildings

Wi-Fi sensors use radio frequency signals instead of light or visual prompts, which allows them to work better in conditions where cameras and lidars are experiencing difficulties. These environments typically include low lighting, changing light, and repetitive environments, such as long corridors. Wi-Fi helps technology achieve the status of a cost-effective alternative to lidars, which are expensive and require a lot of energy.

Initially, Wi-Fi sensors are not aware of the location of the robot and where Wi-Fi access points are located in the environment. As the robot moves, the sensors call the access points and listen to their responses, which are then used as landmarks. The technology is based on the fact that each incoming and outgoing wireless signal carries its own unique physical information, which can be used to determine the location of robots and access points in relation to each other. Algorithms allow sensors to extract the obtained information and make calculations. Sensors continue to gather more information and can eventually locate the robot.

{{quote 'We can use WiFi signals, which are essentially free, to conduct reliable and reliable probing in visually difficult conditions. Wi-Fi sensing could potentially replace expensive lidars and complement other low-cost sensors, such as cameras, in these scenarios, said Aditya Arun, PhD student in the Department of Electrical and Computer Engineering at the Bharadiya Laboratory and first author of the study. }}

The technology was tested on one of the floors of the office building, where several access points were placed around the room. Then the robot was equipped with Wi-Fi sensors, as well as a camera and lidar for comparative measurements. The team controlled the robot and forced him to drive several times on the floor. The robot also turned beyond the corners and passed through long and narrow corridors with a brightly and dimly lit space. Tests showed that the accuracy of localization and mapping provided by Wi-Fi sensors was on the same level as commercial cameras and lidars.^[1]

Notes