The US Army has confirmed its intent to begin testing an autonomous prototype of the Armored Multi-Purpose Vehicle (AMPV) by 2026. This initiative represents a critical step in the service’s push toward integrating autonomy and AI into its future armored formations. Developed by BAE Systems under the Next Generation Combat Vehicle (NGCV) portfolio, the AMPV is being adapted for unmanned operations as part of a broader modernization strategy focused on manned-unmanned teaming and battlefield automation.
AMPV Program Overview and Evolution
The AMPV program was initiated to replace the aging M113 family of tracked armored vehicles across U.S. Army Armored Brigade Combat Teams (ABCTs). The baseline AMPV shares 72% commonality with the M2 Bradley Infantry Fighting Vehicle and is produced by BAE Systems at its York, Pennsylvania facility. It comes in five mission-configured variants—General Purpose, Medical Evacuation, Medical Treatment, Mortar Carrier, and Mission Command—designed to provide improved mobility, protection (under STANAG Level 3), and power generation capabilities compared to its predecessor.
In December 2022, BAE secured a full-rate production contract worth $1.6 billion for up to 586 AMPVs through FY2027. As of mid-2024, over 300 units have been delivered to the U.S. Army. While the initial focus has been on manned operations with enhanced digital architecture and survivability features, recent developments indicate a pivot toward integrating autonomy kits and AI-enabled control systems into select vehicle configurations.
Autonomous Variant Development Timeline
The autonomous AMPV effort is led by the U.S. Army Combat Capabilities Development Command Ground Vehicle Systems Center (CCDC GVSC), in collaboration with BAE Systems under Cooperative Research and Development Agreements (CRADAs). According to statements from Army Futures Command officials at AUSA 2025 and corroborated by Defense News reporting in October 2025, a technology demonstrator featuring basic autonomous functionality is expected to be ready for field testing by Q3 2026.
This prototype will likely incorporate GVSC’s Modular Open System Architecture (MOSA)-compliant Robotic Technology Kernel (RTK), which enables plug-and-play integration of autonomy software stacks such as leader-follower algorithms or waypoint navigation modules. The test platform will initially focus on low-risk logistics missions—such as casualty evacuation or supply transport—before progressing toward more complex tactical roles.
Why Autonomy for AMPV?
The decision to explore an unmanned variant of the AMPV stems from several operational imperatives:
- Force Protection: Reducing crew exposure during high-risk missions like CASEVAC or resupply under fire.
- Manned-Unmanned Teaming: Enabling coordinated operations between manned Bradleys or Abrams tanks and robotic support elements.
- Battlefield Sustainment: Using autonomous AMPVs as logistics mules or mobile command relays within contested environments.
- Platform Commonality: Leveraging existing production lines and sustainment infrastructure while adding autonomy layers without redesigning from scratch.
The Army has already demonstrated similar concepts using optionally-manned versions of other platforms such as the Palletized Load System A-Kit or robotic M113s during Project Convergence experiments in Yuma Proving Ground since 2020.
Technical Challenges Ahead
Despite growing momentum behind robotic ground systems development—including programs like RCV-Light/Medium—the path toward an operationally viable autonomous AMPV remains fraught with technical hurdles:
- Sensing & Perception: Ensuring reliable obstacle detection and terrain classification across varied environments remains difficult without LIDAR redundancy or advanced EO/IR fusion.
- Cybersecurity & EW Resilience: Unmanned platforms are vulnerable to GPS spoofing/jamming and cyber intrusions; hardened comms protocols are essential.
- Crew Override & Fail-Safe Logic: Developing robust human-on-the-loop controls that allow manual intervention when autonomy fails is critical for safety certification.
- Pace of Integration: Retrofitting legacy vehicles like AMPVs with autonomy kits requires extensive validation within existing C4ISR architectures used in ABCTs.
The CCDC GVSC has been investing heavily into simulation-based testing environments such as Hardware-in-the-Loop labs at TARDEC’s Detroit Arsenal campus to accelerate software validation cycles before live trials commence in late FY2026.
Tactical Implications for Future Brigades
If successful, an autonomous AMPV could become a key enabler within future ABCT formations operating under Multi-Domain Operations doctrine. Potential use cases include:
- Tactical CASEVAC under fire: Removing medics from direct danger zones through pre-programmed casualty retrieval routes.
- C4ISR relay nodes: Serving as mobile mesh network hubs when equipped with Line-of-Sight/BLOS comms payloads like TSM-X MANET radios or SATCOM terminals.
- Munitions resupply drones-on-wheels: Delivering ammunition pods autonomously via pre-mapped corridors during high-tempo engagements against peer adversaries.
This aligns with broader U.S. Department of Defense goals outlined in the Joint All-Domain Command & Control (JADC2) framework—where distributed sensors/shooters must be supported by equally agile logistics nodes capable of semi-autonomous operation across denied terrain.
The Road Ahead: From Prototype to Fielding?
The upcoming test phase will be pivotal not only for validating technical feasibility but also for informing future procurement decisions about whether unmanned variants should be fielded alongside crewed AMPVs—or even replace them in certain roles entirely. Lessons learned will likely feed into parallel efforts such as Optionally Manned Fighting Vehicles (OMFV) and Autonomous Transport Vehicles under NGCV umbrella programs funded through FYDP allocations beyond FY27.
No firm timeline has yet been set for productionizing an unmanned AMPV variant post-2026 testing; however, early results may influence upcoming POM cycles if performance metrics meet threshold requirements set by TRADOC Capability Managers for Maneuver Support & Sustainment systems.
