New images have surfaced of a US Army M1 Abrams tank equipped with an experimental artificial intelligence (AI)-driven target detection system designed to speed up how quickly threats can be spotted and engaged.
The images were released to the US military’s Defense Visual Information Distribution Service (DVIDS) website on February 13, 2023. However, they were actually ingested on November 5, 2022 as part of the five-week Project Convergence 2022, or PC22, event that took place in California.
The C5ISR Center partnered with Army Soldiers for five weeks during Project Convergence 22 to test prototypes during force-on-force experiments. DoD
As the official captions note, Army Soldiers have partnered with engineers and scientists at the Command, Control, Communications, Computers, Cyber, Intelligence, Surveillance, and Reconnaissance (C5ISR) Center to test prototypes of the technology used as part of the Advanced Targeting and Lethality Aided System (ATLAS) program is being developed. “A wide range of supported target acquisition, tracking and reporting functions [were tested] …in a realistic combat environment” during PC22, read the captions.
We can see components of ATLAS being tested in the images. Most notable is the box-shaped sensor unit, mounted on a rotating base on the M1’s turret just behind the 120mm main gun.
A closeup of the ATLAS sensor tower. DoD
A black box can be seen under the main sensor circled in red below, and similar boxes can be found elsewhere around the tank. These appear to be set up for the exercise as part of the Instrumentable-Multiple Integrated Laser Engagement System Combat Vehicle Tactical Engagement Simulation System (I-MILES CVTESS) and are therefore not associated with the ATLAS system. I-MILES is used to detect and evaluate hits with lasers to simulate combat and evaluate combat damage.
DoDI MILES CVTESS boxes seen elsewhere on the tank. DoD
There’s also some sort of another sensor on the front of the Abrams. The exact nature and purpose of this sensor remains unclear, although it doesn’t appear to be standard on other M1s we’ve seen.
DoDDoD
On the back of the Abrams we see the main sensor unit from behind, as well as other pieces of equipment.
DoDA rear view of the Abrams. DoD
Specifically, circled in red on the lower left is a large black box that is part of the I-MILES laser system. We also see what appears to be the back of an air conditioner, likely for the various computers needed to run and drive ATLAS, also circled in red.
DoD
ATLAS, a joint initiative of U.S. Army Combat Capabilities Development Command (DEVCOM) C5ISR and Armaments Centers, “uses cutting-edge sensor technologies and machine learning algorithms to automate manual tasks during passive target acquisition, allowing crews to engage three targets at the same time it would normally take until they attack you,” according to DVIDS.
As C4ISRNET’s Nathan Strout points out, ATLAS starts with the tank-mounted optical sensor feeding images of the environment into AI algorithms for object detection and image classification. From there, images of the detected threats are displayed on the tank’s touchscreen interface for tank commanders to view along with weapon/ammo selection options.
The program essentially aims to speed up target detection by experimenting with and implementing AI, eliminating the need for tank crews to rely solely on manually locating targets.
This all requires a human-in-the-loop, it should be noted that tank commanders still have to choose which targets to fight and with which weapon and ammo types. like dr John Graybeal, senior engineering psychologist at C5ISR’s Human Perception Laboratory, stated in an Army article back in 2021 that ATLAS is designed to support manual target recognition, increasing the chances of engaging enemy targets. “ATLAS uses an assisted target detection system that scans a field of view so the soldier is assisted by an artificial intelligence system that is also searching the crime scene.”
The Army’s effort to improve tank lethality by implementing machine learning and AI technologies has been in the works for several years. Back in July 2020, the service announced that engineers at Picatinny Arsenal, New Jersey, are developing the Advanced Lethality and Accuracy System for Medium Caliber (ALAS-MC), an armor system that will include “a medium-caliber weapon, ammunition, fire control and sensors, to effectively engage targets at longer ranges.” The ALAS-MC armament system incorporated a sensor unit with a 50mm autocannon known as the XM913 (see below).
ALAS MC. United States Army
Then, in October 2020, the ATLAS system was demonstrated by the Army on a General Dynamics Land Systems Griffin I test bed vehicle. Here, ALAS-MC’s 50mm self-loading cannon was paired with an Infrared Assisted Target Recognition (AiTR) sensor sphere, with images processed by AI algorithms and sent to a user interface.
Aided Target Recognition (AiTR) sensor and autoloading ALAS-MC gun demonstrated by the US Army in 2020. US Army
Of course, putting ATLAS into service could have huge potential benefits for the Army. AI and computer-aided targeting would not only allow tank crews to eliminate threats faster – it could also be used to detect threats that humans cannot. This could also be expanded to potentially help tank crews prioritize which threats need to be neutralized first in certain scenarios. Recently we’ve seen Panoramic Infrared Search and Track Type Systems (IRST) used on ships with capabilities similar to ATLAS. Similar systems could be used to provide sustained 360 degree coverage for emerging threats to land vehicles.
AI could also help with land-vehicle-based Distributed Aperture Systems (DAS), a likely feature of the future AbramsX main battle tank, which will allow crews to “see” through the tank’s hull with augmented reality goggles. Both ATLAS, 360 IRST and DAS-like systems could, with the right software, digitally highlight objects of interest or threats in their environment. In fact, it’s possible that a common AI-driven software architecture could be deployed across various electro-optical and infrared ground vehicle situational awareness systems.
Additionally, modern main battle tanks, particularly the state-of-the-art M1 Abrams, require rigorous and lengthy periods of training for crews to learn how to operate them. The use of ATLAS for improved target selection and target engagement would allow tank units to better balance the demands required for tanks to be used successfully in combat. As we highlighted in this recent War Zone feature, tankers must master a variety of challenges during tank-to-tank combat, including tactics, vehicle operation, and weapon use.
And while ATLAS is focused on human-in-the-loop contexts, the new technology could also be a stepping stone to targeting capabilities desirable for larger unmanned ground vehicles. All of this is happening amid a broader push by the DoD toward AI and autonomous weapon systems, and is just one example of how AI is an increasingly important component of future weapon development.
