DARPA has launched the Robotic Autonomy in Complex Environments with Resiliency (RACER) program, which aims to develop new algorithms for autonomous combat UGVs (unmanned ground vehicles) that will allow them to meet or exceed soldier driving abilities.
Most autonomous self-driving vehicles have been designed to operate well-structured and highly predictable environments. So far, in complex militarily-relevant settings, robotic vehicles have not demonstrated operationally relevant speed and are not autonomously reliable. While unmanned vehicle platforms that can handle difficult terrain exist, their autonomy algorithms and software are often unable to process and respond to changing situations well enough to maintain necessary speeds and keep up with soldiers on a mission. The RACER program aims to make sure that algorithms are not the limiting part of these systems.
Over a period of four years, RACER will develop new algorithm technologies that maximize utilization of the sensor and push the mechanical limits of UGVs. These algorithms will be continually field-tested at DARPA-hosted experiments across the U.S. on a variety of terrain. DARPA will provide advanced UGV platforms that research teams will use to develop autonomous software capabilities through repeated cycles of simulations and tests on unstructured off-road landscapes. The program will also develop simulation environments that will support rapid advancement of self-driving capabilities for future UGVs.
Stuart Young, program manager leading the RACER project, commented: “In order to achieve RACER goals of increased speed and resilience, we need to embrace learning approaches that automatically tune system parameters in real time. Successful software will extract features from sensor data and use that information to make on-the-spot driving decisions.”