The US Army is preparing to conduct its first live-fire test of the Robotic Autonomy in Complex Environments with Resiliency (RACER) breacher—a next-generation autonomous ground system designed for high-risk minefield clearance. Scheduled for 2025 at Yuma Proving Ground, this marks a significant milestone in integrating artificial intelligence and autonomy into combat engineering missions.
RACER Program Origins and Objectives
The RACER program was launched by the Defense Advanced Research Projects Agency (DARPA) in 2020 with the goal of enabling off-road autonomous navigation at operationally relevant speeds. Unlike traditional UGVs that rely on pre-mapped environments or GPS-based waypoints, RACER vehicles are designed to perceive and adapt to unfamiliar terrain using onboard sensors and machine learning algorithms.
Initially focused on reconnaissance and mobility applications, the program has evolved to include more complex mission sets such as breaching operations. The RACER Heavy Platform variant—developed by a team led by Carnegie Mellon University’s National Robotics Engineering Center (NREC)—is now being adapted for explosive hazard clearance roles under a cooperative effort involving DARPA and the US Army Combat Capabilities Development Command (CCDC) Ground Vehicle Systems Center.
Autonomous Breaching: A High-Risk Use Case
Minefield breaching is among the most dangerous tasks for combat engineers. Traditionally requiring manned vehicles like M1150 Assault Breacher Vehicles or manually emplaced line charges such as MICLICs (Mine Clearing Line Charge), these operations expose personnel to direct fire and explosive threats. An autonomous solution could significantly reduce risk while accelerating tempo during offensive maneuvers.
The RACER breacher prototype integrates a suite of sensors—including LIDAR, EO/IR cameras, radar, and inertial navigation systems—to autonomously detect terrain features and potential obstacles. For mine detection specifically, it is expected to incorporate ground-penetrating radar (GPR) or electromagnetic induction sensors—though exact sensor configurations have not been publicly disclosed.
In addition to detection capabilities, the platform must demonstrate sufficient traction control and path-planning logic to navigate through soft soil or craters caused by detonations. The upcoming live test will evaluate whether its autonomy stack can handle such conditions without operator intervention.
Yuma Proving Ground Test Parameters
The planned 2025 demonstration at Yuma Proving Ground will be the first time an autonomous vehicle attempts live minefield clearance without teleoperation or remote piloting. According to DARPA officials cited in recent briefings at AUSA 2024, the test will involve a simulated obstacle belt containing both inert training mines and live explosive devices under controlled conditions.
Key evaluation criteria include:
- Autonomous navigation through cluttered terrain with minimal GPS availability
- Detection and avoidance—or neutralization—of buried explosive hazards
- System resiliency after blast exposure or partial degradation
- Mission completion time compared to manned equivalents
If successful, this would represent a significant leap forward not only for robotic breaching but also for broader applications of AI-enabled uncrewed ground systems (UGS) in contested environments where EW threats may deny communications or GNSS signals.
Platform Design: From Racer Vehicle Base to Combat Engineer Tool
The RACER Heavy platform is based on a modified Polaris MRZR Alpha chassis but significantly up-armored and reconfigured for autonomy. It features modular payload bays that can support various mission kits—including breaching plows or rollers similar to those used on Abrams tanks equipped with mine-clearing blades.
A key design feature is its autonomy software stack developed under DARPA’s RACER-Sim initiative—a simulation-based training environment that allows developers to iterate AI behaviors before real-world deployment. This enables rapid adaptation of navigation strategies based on environmental feedback rather than hard-coded logic trees.
The final operational version may also integrate CBRN protection sensors or deployable counter-IED payloads depending on mission needs. Powertrain modifications have been made for low acoustic signature operation—important when operating near enemy lines during shaping operations ahead of main assaults.
Tactical Implications and Future Roadmap
If proven viable during testing at Yuma Proving Ground, the RACER breacher could be transitioned into an Army Program of Record as early as FY2026 under Next Generation Combat Vehicle initiatives. Its utility extends beyond conventional warfare; such platforms could be invaluable in urban clearance scenarios where booby traps complicate movement corridors.
DARPA’s broader vision includes swarming concepts where multiple UGS platforms cooperate autonomously across a battlespace—sharing sensor data via mesh networks while executing distributed tasks like route clearance or logistics resupply under fire. The RACER program serves as an enabling technology base for these future force structures.
Challenges Ahead: Autonomy Under Fire
A major hurdle remains ensuring robust autonomy under adversarial conditions—including GPS jamming/spoofing, cyber intrusion attempts against onboard processors, and physical damage from blasts. While DARPA has made strides in resilient autonomy frameworks through programs like OFFSET and Gremlins-XAI integration layers, real-world performance remains unproven at scale.
The Yuma trial will provide critical data not only on technical feasibility but also inform doctrinal changes regarding how future engineer units might be structured around robotic assets rather than traditional sapper teams alone. Integration with existing command-and-control architectures—and deconfliction with manned maneuver elements—will require careful planning if such systems are fielded broadly.
Conclusion
The upcoming test of the RACER autonomous breacher marks a pivotal moment in military robotics development—shifting from lab-based trials toward realistic battlefield application. Success could redefine how armies approach one of land warfare’s most perilous challenges: clearing lanes through lethal obstacles without risking human lives at every step forward.
