 |
| The US Army Uses AI Technology on the M1 Abrams Tank |
Washington - The United States Army conducted trials using artificial intelligence (AI) technology on the M1 Abrams tank targeting system. This program aims to accelerate target recognition and reduce the reliance on tank crews in manually aiming targets.
Testing of AI technology on the M Abrams tank was released via the US military's Defense Visual Information Distribution Service (DVIDS) website on February 13, 2023. However, the image was actually taken on November 5, 2022, as part of the five-week Project Convergence 2022 event, or PC22 , which took place in California.
This program is referred to as the Advanced Targeting and Lethality Aided System (ATLAS) Program. The ATLAS program was developed in partnership with the US Army's Command, Control, Communications, Computing, Cyber, Intelligence, Surveillance, and Reconnaissance (C5ISR) Center.
"This prototype was developed to enhance a wide range of target acquisition, tracking and reporting capabilities in a realistic combat environment," said DVIDS.
The ATLAS component tested on the M1 Abrams tank was a box-shaped sensor unit mounted on a rotating base in the tank's turret, immediately behind the 120mm main gun. It is part of the Combat Vehicle Tactical Engagement System Multiple Instrument-Integrated Laser Engagement System (I-MILES CVTESS).
I-MILES is used to detect and score hits using lasers to simulate combat and assess the damage done. The front of the M1 Abrams tank also uses another type of sensor. The exact nature and purpose of this sensor remains unclear, although it does not appear to be standard on the M1 Abrams tank. At the rear of the M1 Abrams tank, there is the rear main sensor unit as well as other equipment.
The way ATLAS works begins with an optical sensor mounted on a tank to input images of the surrounding area into an AI algorithm, for object detection and image classification. From there, an image of a detected threat object is displayed on the tank's touchscreen interface for the tank commander to view, along with options for selecting which weapon/ammunition to use.
“Using state-of-the-art sensing technology and machine learning algorithms, aims to automate manual tasks during passive target acquisition. This technology allows the crew to engage three targets at once," said DVIDS.