The UK Defence Science and Technology Laboratory (Dstl) has conducted an artificial intelligence (AI) platform recognition trial.
It aimed to build datasets and validate algorithms to support the rapid development of AI-based defence applications and gradually boost weapon capabilities while minimising development costs.
According to Dstl, the lack of datasets for any task, such as target recognition, is one of the key challenges that defence suppliers commonly face.
The trials were therefore conducted to facilitate the safe and reliable transmission of data by using different platforms of interest such as infantry fighting vehicles, tanks, and armoured personnel carriers.
Operated/commanded by Dstl’s staff, the platforms were deployed at five different areas of Dstl’s trial range located in the Salisbury Plain, England.
As part of the trials, various industry partners, including Lockheed Martin, Thales, Frazer-Nash Leonardo, and MBDA, successfully recorded the data from a wide range of sensors and cameras and transferred it to the Ministry of Defence (MoD).
The MoD will use this dataset to support the development of AI and machine learning (ML) techniques, as well as target recognition algorithms to enhance overall battlefield capability via Dstl’s Cooperative Weapons technology demonstrator programme.
In addition, the companies were also able to retain access to data that can further be leveraged to train and develop the ML algorithms.
All the trials were hosted by Defence Equipment and Support (DE&S) on the Defence Digital Cloud. DE&S also manages data gathering contracts.
Dstl Weapon Systems programme manager Simon Zavad said: “The trial drew on specialist
expertise across multiple areas of Dstl, bringing them together to collaborate with industry partners and government colleagues.
“The data gathered could accelerate development of AI software, improving accuracy of data to enable commanders to make better combat decisions.”