Autonomous research systems have the potential to revolutionize the way research is done by using artificial intelligence to drive robotic experiments over hundreds or thousands of iterations. Already, autonomous research robots such as ARES™ have learned to grow carbon nanotubes at controlled rates and optimized the structure of a 3D-printed twisted barrel to maximize energy absorption. However, to date, building a research robot has required significant investments in software development.
ARES OS™ has been freely available since the fall of 2020 from the Air Force Research Laboratory as open source software that researchers can use as a foundation to build their own research robots. With roots in autonomous robotics, ARES OS™ is architected to lower the barrier to entry to autonomous research systems by structuring hardware/software interface modules, analytical modules that provide feedback information, and planning modules that can use AI and machine learning to direct iterative experiments. ARES OS™ can save months or years of time compared to custom code.
Several groups are evaluating ARES OS™, including in industry, academia, and government laboratories. ARES OS™ has been successfully deployed on three autonomous research robots to date, including for carbon nanotube synthesis, flow chemistry, and additive manufacturing.