Updated: Jul 14, 2022
Making the right decision at the right time necessitates real-time situational awareness, and decision support systems. Sponsored by Omnisys
Operational Mission Management
As the complexity of operational missions and systems keeps growing, making the right operational decisions at the right time necessitates (1) real-time situational awareness, and (2) decision support systems, providing automated recommendations. These two capabilities are the core of operational mission management systems (MMSs).
The situational status provided by MMSs encompasses information relating to the present state of our forces and the battlefield surrounding them, within the context of the operational mission. Situational status updates may include, for example:
The state of our forces and systems, as well as those of friendly and enemy forces;
Various mission performance parameters relating to our forces and systems, in terms of past performance as well as estimated current and future performance;
Anomalies in the behavior of our systems, which may result from malfunctions, interference or mutual interference, weather conditions, and so forth;
Intelligence updates relating to the enemy's forces and systems;
Anomalies in enemy activity, including its concept of operations (CONOPS) and system parameters.
The above-mentioned situational status is typically derived from the analysis of data from a wide variety of information sources, such as:
Sensor outputs, including radars, signal intelligence systems, optic/electro-optic sensors;
acoustic sensors, and so forth;
Command and control (C2) outputs;
Human intelligence information.
This may be performed using both classic methods, such as heuristic algorithms and state-of-the-art methods, such as machine learning (ML) and artificial intelligence (AI) algorithms.
Automated recommendations are produced by MMSs either in accordance with a priori intelligence information (before the mission starts) or in response to situational status updates during the mission. Focusing on actions our forces and systems can take, these recommendations may comprise, for instance:
Force deployment changes;
Platform placement / trajectory changes;
System configuration updates, e.g., frequency allocation revisions;
System resource re-allocation, i.e., changes in the allocation of sensors, effectors;
transmitters, or platforms to targets and areas / volumes of interest.
The decision support component of MMSs is often based on a combination of simulations and optimization tools. The simulations are used to estimate what would happen in various cases, and the optimization process determines which cases should be examined and selects the most suitable configuration.
Additionally or alternatively, the decision support component of MMSs may employ logic algorithms, yielding suggested responses for a set of predefined events. For instance, if a ground-based radar measures an increase in noise level at a certain spatial region and a specific frequency range, some actions may be taken to alleviate this problem even without specific information about the interference source.
These actions may include, e.g., changing the radar's transmission frequency, adjusting its search pattern, or disabling track initialization for certain angular swaths. The logic algorithms may utilize methods known in literature such as decision trees and random forests.
Most currently available MMSs address a single mission. However, multi-mission MMSs (MM-MMSs) can be operational performance multipliers as they manage resources that are shared by several missions.
The most prominent shared resource is the electromagnetic spectrum, which is used by any system that transmits or receives electromagnetic radiation. Such systems include wireless communication systems, global navigation satellite system (GNSS) receivers, radars, signal intelligence systems, electronic warfare systems, and so on.
Most operational missions today rely heavily on wireless communication networks and/or on GNSS receivers, and are therefore susceptible to mutual interference, jamming, or spoofing. The only way to assure continuous coexistence is real-time spectrum management, taking into account all spectrum-dependent systems in the region, which may be used for several missions.
Another example of a shared resource is a military transport vehicle, which can serve different units at different locations but can only have a single path at a given time.
The main technological challenge in MM-MMSs is multi-mission optimization. Multi-mission optimization is characterized by:
A very large number of degrees of freedom that must be simultaneously optimized, resulting in high computational complexity and memory requirements.
Diversity in the context of the different degrees of freedom. Classic optimization algorithms typically deal with a single type of problem, such as deployment, trajectory planning, or resource allocation. Multi-mission optimization is often required to address two or more types of problems together.
Moreover, defining the target function is especially difficult for multi-mission optimization. The optimization target function determines what the optimization process tries to optimize, by providing a performance score for each examined scenario.
For MM-MMSs, the target function may take into account issues such as mission priority and mission mutual effects. To obtain meaningful results, the target function is typically calibrated by extensive operations research. Reinforcement learning may also be used to adaptively tailor the target function to the needs of the operational teams.
AI-Based Situational Awareness
As described above, situational awareness may be obtained with the use of AI algorithms.
Within the context of MMSs, the AI algorithms must support model training using small datasets, to allow flexibility for responding to quick changes in the battlefield. Furthermore, the algorithms should be based on minimal assumptions, to reduce model mismatch and yield accurate results. Common AI algorithms usually adhere to only one of these requirements. For instance, deep neural networks (DNNs) require very little a priori information but necessitate large training datasets.
In contrast, Gaussian mixture models (GMMs) and support vector machines (SVMs) are based on specific model suppositions, but small training sets are often sufficient for them.
One of the proprietary suites of algorithms developed by Omnisys, a leading provider of mission optimization solutions for defense and homeland security, is called "agile statistical modeling (ASM)." It was developed with the above-mentioned requirements in mind. Moreover, ASM is a white-box method, i.e., its models can easily be visualized and manually adjusted. This attribute is of great importance when dealing with strategic missions, where errors are often unacceptable, and the ability to estimate in advance the system's performance in various scenarios is critical.
MMSs are gradually becoming crucial for the successful completion of operational missions in the modern battlefield, which is characterized by rapid changes and a myriad of systems addressing numerous operational requirements. AI-based situational awareness and multi-mission optimization are two key drivers in the continued development of these systems.
Omnisys is a global provider of leading mission optimization solutions for defense and homeland security. The Omnisys BRO suite of battle resource optimization systems enables end-to-end planning, management, and debriefing for integrated land, sea, Spectrum and air missions. The mission optimization leader in Israel’s defense industry for more than two decades, Omnisys delivers combat-proven solutions for diverse critical missions including intelligence gathering, spectrum management, air defense, air surveillance, border security and electronic warfare. Possessing deep multidisciplinary expertise, Omnisys develops cutting-edge analysis and optimization technologies. Omnisys empowers mission commanders and other users to optimize critical decisions for maximized mission performance.
Learn more about the BRO cutting-edge solutions at the Omnisys website.
**Sponsored by Omnisys