L

The Indian Air Force became the first among the services to establish a dedicated center of excellence for the artificial intelligence. In the context of military modernization, such digital tools are radically redefining the technological elements and such gaps are widely discussed. Artificial intelligence is rapidly changing supply logistics, intelligence gathering, sensors, and military robots through the application as commonly knows as the Internet of Battlefield Things(IoBT).

The IAF Centre of Excellence for Artificial Intelligence under the aegis of UDAAN (Unit for Digitisation, Automation, Artificial Intelligence and Application Networking) was inaugurated by Air Marshal Sandeep Singh, Vice Chief of the Air Staff (VCAS), on 09 July 2022 at Air Force Station Rajokri, New Delhi.

A Big Data Analytics and Al Platform has been commissioned in the IAF’s Al Centre, for handling all aspects of Analytics, Machine Learning, Natural Language Processing, Neural Networks and Deep Learning algorithms. The high-end compute requirements would be undertaken by the latest Graphical Processing Unit powered servers.

Addressing the gathering, VCAS said that IAF has taken proactive steps to embed Industry 4.0 and Al based technologies in its war fighting processes. He reiterated that the AI COE with high end compute and big data storage capabilities, coupled with full spectrum Al software suites, would substantially enhance operational capability of IAF.

According to the officials, the Al based applications are being developed with inhouse expertise in coordination with various PSUs, MSMEs and leading academia in the field of Artificial intelligence.

Artificial Intelligence(AI) in Air Combat

The next generation combat jets is largely software-centric for its combat operations. The elements of engagement are based on the data for the target detection, tracking and combat operations. In the case of only operational fifth generation fighter jet–F-22 and F-35– the target engagement activities are heavily software-generated, covering upto 95% the entire combat activities. The cockpit is so filled with sensors and computer processors that Algorithms are already flying planes. The data received from the multi sensors-sub-systems forms the basis of such engagement. However that is also the biggest challenge for a pilot to read through and process the high-speed data in a complex air combat environment. Also the Multi-Platform Multi-Sensor Data Fusion (MPMSDF) creates real time awareness, integrating battlespace surrounding for the decision making.

This is the core of AI based Decision Support Systems (DSS) which can fundamentally address such complex environment and analyze the sensory data through processors. Specific to DSS, India does have foundational base as Defense Research and Development Organization (DRDO) has initiated and developed framework.

DRDO’s Centre for Artificial Intelligence & Robotics (CAIR) has established a decision support system (DSS) framework, which is completely driven by knowledge base maintained as ontologies. Algorithms like Multi Criteria Decision Making (MCDM), Swarm Algorithms, Game theoretic approaches towards resource allocation, Search algorithms etc. built on the distributed framework of Hadoop, support in giving “intelligent” and informed solutions. This framework has been used for providing Decision Support for Deployment, Transport Allocation, Convoy composition and scheduling and Coastal Surveillance.

While the algorithm is already flying planes, AI is now more about how to fight its own, and how to get pilots to trust the AI. It zeroes in on the OODA loop—the decision cycle of orient, observe, decide, and act. What are the fundamental basis for AI and OODA loop? In a well chronicled dogfights where American fighter aircraft F-86 won against the Russian MiG-15, the victories were based on the fact that American pilots had a much shorter OODA loop in the observing and acting phase of combat operation. The F-86 had the much wider field of vision and easier hydraulic controls that allowed them to outrun the OODA loops of Russian pilots. So, the role of AI is about processing such information which is difficult to achieve in shorter time-frame. AI can drastically reduce the OODA loop. This is one of such applications.

Further, AI is also being deliberated across the C4I (Command, Control, Communication, Computers and Intelligence) management system. The research in this area is taking place at a staggering speed to achieve a MIL Grade standard and using AI in a SaaS (Software as a Service) model.

AI will be crucial for the AMCA program– Advanced Medium Combat Aircraft, which is to be designed and developed as a fifth-generation medium-weight fighter aircraft. The level of autonomous functionality will be based on sixth generation technologies, especially the smart wingman concept and optionally manned/unmanned combat platform. Such concept is heavily based on the AI. AI is driving the navigational systems for such complex teaming exercise in conjugation with integrated cockpit with the multi-sensor data fusion and multiple tactical data links. How much AMCA will incorporate such critical AI-enabled technologies? Certainly, It will depend on how do we leverage AI.

AI for Autonomous Vehicle

AI is the reason behind the explosive growth in innovative applications of Unmanned Aerial Vehicles (UAVs) and its military applications. AI applications in UAVs are spreading across an impressive variety of domains, including ISR (intelligence, surveillance and reconnaissance) and Targeting. Worldwide, military is heavily spending on the computer vision capability of artificial intelligence for activities like detect and hunt down submarines, detect an enemy intrusion, or decode messages using machine learning abilities.

AI application is also playing crucial role in developing anti-combat drone solutions. In a recent development, a US Start-up Epirus has developed Leonidas, a technology that can disable a hostile drone while leaving a friendly drone a few feet distant unharmed. Using super-dense Gallium Nitride power amplifiers and based on the AI algorithms, Leonidas uses direct energy to precise frequencies which can take out both large fixed-wing drones and small quadcopters. 

India’s ambitious unmanned projects– Rustom – II UAV and Ghatak UCAV– are under development phase. Such Intelligent Unmanned Systems is based on Computer Vision Processing and Artificial Intelligence. Ghatak UCAV is a stealth unmanned combat aerial vehicle (UCAV) initiated under the program– AURA( Autonomous Unmanned Research Aircraft). AI- enabled system will be critical in the development of the target detection and use of Lethal Autonomous Weapon Systems (LAWS).

It is also important to highlight that while the Indian Air Force (IAF) squadrons are depleting, it is imperative that UAV capabilities get the thrust and support for developing indigenous UAV projects.

Besides, IAF has been pioneering AI is area of aircraft maintenance. IAF has substantially digitised its fleets onto the electronic maintenance management systems. IAF has also digitized the entire inventory management system which works on on AI-based formulation to come out with predictive maintenance or predictive threat scenarios or red flags.

Challenges for AI

While AI is evolving, some of the challenges are critical especially in its military applications. Here is the need to address some questions to this effect- whether algorithms can be trained to effectively execute mission planning behaviors in unpredictable scenarios; can machines be taught combat strategies; could sufficiently generalized representations be built to capture the richness of the planning problem itself across the threat matrix.

“The answer to these questions will help us firm up our requirement specifications that will essentially be a starting document vis-à-vis the expected outcomes. If we tend to utilise AI heavily in combat aviation, we may need to redefine or even abandon certain traditional principles,” said former IAF Chief RKS Bhadauria.

AI is not just a tactical advantage but it has become a necessity. Nations such as China, US and Russia, and many others are already investing heavily in AI. The need for the big data to train and test combat systems will be required for IAF. The data collection, assimilation, and analysis together will drive the AI for the next generation. Integrating AI in military strategies will be the cornerstone of the defence sector. AI is the only way to navigate through this new paradigm of warfare.