Listen to this article
GE Research successfully completed year one of a project with the U.S. Army through its Scalable Adaptive Resilient Autonomy Program (SARA). The goal is to develop a risk-aware autonomous ground vehicle that can navigate complex, off-road test conditions.
GE’s project was one of eight funded by the U.S. Army’s Combat Capabilities Development Command Army Research Laboratory to advance autonomous, off-road navigation capabilities for military ground vehicles. The Army is committed to enhancing autonomous vehicle capabilities to perform safely and reliably in complex battlefield environments such as wooded areas where paths are unstructured, conditions uncertain and for which no maps exist to provide direction.
“Being able to essentially go into a new space, assess where you are, what you are looking at, understanding the uncertainty with which you’re operating under and then behaving accordingly is essentially what we’re driving toward with the SARA program,” said John Lizzi, GE Research Lab’s robotics and autonomy tech leader.
GE said a key factor enabling the breakthrough in addressing risk was the integration of GE’s Humble AI technology. Humble AI is an algorithmic approach that GE claims is capable of taking a step back and “assessing much like a human” what to do when it encounters an uncertain situation.
Here is a video that shows GE’s autonomous robot, named “ATVer,” navigate around trees and other obstacles in the woods at GE’s upstate New York campus.
“In our autonomous ground vehicle system, Humble AI was the key mechanism that enabled it to decipher known paths vs. uncertain paths when navigating unstructured, off-road terrains,” said senior robotics scientist Shiraj Sen. “It processed the information gathered from the eyes of our vehicle – the vision and LIDAR technologies – and was able to move safely and avoid objects like tree or brush that might stop it in its tracks.”
GE had already field tested its Humble AI algorithms to safely optimize the control of wind turbines to maximize energy input. In the case of wind turbines, the Humble AI operates within a zone of competency where it bases its decisions on known operating scenarios with which it is familiar. When it encounters a scenario it has never seen, it is designed to take a step back and relinquish control of the turbine into a default safe mode.
“In future Army scenarios, autonomous systems will have to reliably plan in the presence of challenging features they encounter while maneuvering in complex terrain,” said SARA program manager Eric Spero. “Incorporating risk and uncertainty into the autonomy decision-making process enables our testbed platforms to show us what it looks like to plan a direct path instead of taking the long way around.”
“Our project and partnership with the US Army has really enabled us to make some important advances in autonomous systems,” Sen said. “We believe the advances made on this project will not only help accelerate the deployment of future driverless vehicle technologies; they will help encourage more autonomous solutions in other industry sectors like energy, aviation and healthcare that people depend on every day.”
GE also recently conducted a real-world test of its tunneling robot. Using its pneumatic artificial muscles, the prototype autonomously dug a 10 cm tunnel at GE’s campus, sliding past small rocks and other obstacles.