Print Friendly, PDF & Email

The research agency of the US Department of Defense DARPA (Defense Advanced Research Projects Agency) is developing autonomous controls for unmanned combat vehicles in the RACER program (Robotic Autonomy in Complex Environments with Resiliency). As part of the project, technologies are to be developed with which the vehicles can maneuver in difficult terrain at the limit of the performance of their mechanical systems. The DARPA describes its requirements as aiming for speeds that are at least as high as those of manned vehicles.

„RACER soll die Integration und den Einsatz von autonomen Roboter-Kampffahrzeugen in der Armee, im Marine Corps und bei den Spezialeinheiten entscheidend voranbringen“, sagte Stuart Young, RACER-Programmmanager im Tactical Technology Office der DARPA.

As the launch vehicle, dubbed the RACER Fleet Vehicle (RFV), for Phase 1 of the project, Carnegie Robotics has equipped a Polaris RZR S4 1000 Turbo drive-by-wire platform with 360° range and image sensing. Sensors include multiple LIDARs, stereo camera pairs, color and infrared cameras, RADAR, event sensors, and inertial measurement sensors. The data collected by the sensors are stored and processed in a protected, shock and vibration resistant and thermally controlled electronics box. For the task as an RFV, the 74 kW RZR was also equipped with a roll bar and a 7 kW generator to supply the electronics.

Of the four RFVs built for RACER to date, one is currently collecting four terabytes of sensor data per hour to support the artificial intelligence and machine learning-based autonomy algorithms and stack approaches required for rapid maneuvers in complex terrain. Another four vehicles should be available for field trials in March 2022.

To further support software development, DARPA has also collected over 100 terabytes of RFV-based sensor data from more than 500 kilometers of terrain on the east and west coasts of the United States. This data is shared with the teams and managed within a RACER development tool to ensure efficiency and security.

Im Rahmen des RACER-Programms wurden außerdem zwei Aufträge zur Entwicklung von Simulationsumgebungen und -fähigkeiten vergeben, die die Entwicklung von Algorithmen für autonomes Fahren im Gelände ermöglichen. „Diese Simulationsumgebungen werden es den Teams ermöglichen, Teile ihrer autonomen Systeme zu testen und zu validieren, ohne viel Zeit und Geld für Feldtests aufwenden zu müssen“, so Young.

DARPA entwickelt autonome Hochgeschwindigkeits-Geländewagen

The two main customers for the RACER SIM program are Duality Robotics and Intel-Federal.

The three institutes (Carnegie Mellon University, NASA Jet Propulsion Laboratory and University of Washington) were each commissioned to develop an autonomous control system for the RFV, with which the vehicle can independently determine its route in the terrain and reach the specified target point at maximum speed .

Phase 1 will culminate in field trials in March 2022 with the three teams' vehicles at the National Training Center in Fort Irwin, California. There, the vehicles will prove their performance on routes with different terrain and over distances of up to five kilometers.

After the successful completion of phase 1, it is planned to continue the investigations with a combat vehicle demonstrator in the ten-ton class in phase 2. The aim is to achieve a speed and mobility that should be above those of the RFV. In addition, tactical requirements for the new platforms are to be examined.

Editors / gwh