A big problem for autonomous robots is the challenge of navigating large environments that are void of a global positioning system (GPS).
Suman Chakravorty, associate professor in the Department of Aerospace Engineering and the Estimation Decision and Planning Lab at Texas A&M University, and his doctoral student, Saurav Agarwal, have developed an indoor mapping technology that allows autonomous robots to be deployed in extremely large-scale environments, such as Amazon warehouses that are millions of square feet in size, at an affordable cost.
“We believe our navigation technology can revolutionize the field of autonomous robotics and speed up the time-to-market for safe and reliable commercial robots,” Agarwal said.
Chakravorty and Agarwal discovered a major gap in existing automation for warehousing and logistics, and received a $50,000 Innovation Corps grant (I-Corps) from the National Science Foundation to research commercial needs for their technology. Their goal for the research is to develop tools and methods to enable robust long-term autonomy for mobile robots.
This I-Corps project is a result of research into the problem of Simultaneous Localization and Mapping (SLAM). In SLAM, a robot is not given prior knowledge of its environment; it must use its sensory data and actions to simultaneously build a map of its environment and position itself within its uncertain surroundings. Competing methods in this area exhibit positioning errors that may be unsuitable for long-term navigation.
Chakravorty and Agarwal’s mapping and navigation technology achieves three times more robustness in real-world and simulated tests (99.99% success rate in testing) compared to existing state-of-the-art technology. They have shown that their algorithms allow robots to map environments in cases where currently available mapping technology fails.
Material-handling robots that move goods, such as boxes and pallets, in large warehouses and distribution centers are examples of robots that navigate in indoor environments. The mapping technology the team has developed enables robots to navigate safely and reliably, thereby reducing labor costs for warehouse operators and reducing injuries caused by human drivers of forklifts. This will, in turn, enable cheaper and faster access to goods for consumers.
This autonomous navigation technology will enable systems to robustly operate in uncertain environments without GPS, and commercialization of the technology has the potential to revolutionize space exploration, self-driving cars, unmanned aerial vehicles (UAVs) and other such systems that need accurate position estimation.
A key advantage of this project’s technology is enhanced cybersecurity because it does not rely on external signals for navigation. Furthermore, this project will contribute open-source software to the scientific community. It is envisioned that development of a software toolbox that integrates with the popular ROS (Robot Operating System) library will allow researchers to simulate autonomous navigation without GPS.
In less than a year, Chakravorty and Agarwal have taken the idea from a mathematical concept to a proof-of-concept robot running in a live warehouse. The technology has never failed in practice, whereas competing technologies fail 50-60 percent of the time.
The main research challenge for Chakravorty and Agarwal is deploying this technology at scale in real-world warehouses, factories and distribution centers. They are exploring opportunities with the Autonomous Systems Lab under the direction of Srikanth Saripalli, associate professor in the Department of Mechanical Engineering at Texas A&M, to deploy this technology with a real-world forklift.
Two National Science Foundation projects were instrumental in the development of this technology: the projects studied robust estimation and motion planning under uncertainty, paving the way for the robust autonomy solution developed by the two researchers.
This story by Jan McHarg originally appeared on the College of Engineering website.