Army researchers reduced the unpredictability of current reinforcement learning policies so they are more applicable to ground robots. These learning components will permit autonomous robots to reason and adapt to changing conditions. | Credit: U.S. Army
Multi-domain operations (MDO) the Army’s future operating concept, requires autonomous agents with learning components to operate alongside humans. New Army research reduces the unpredictability of current training reinforcement learning policies so that they are more practically applicable to physical systems, especially ground robots.
The post How the Army is improving reinforcement learning for robots appeared first on The Robot Report.
