Artificial intelligence could soon be helping the military predict when equipment will break, defense against attacks, and prevent ships from colliding with one another.
The Pentagon has made no secret that it wants to pair humans with machines to help them make decisions faster. The military has been looking for ways to automate intelligence processing using a new tech warfare cell.
Already, these algorithms are being used by commercial firms to forecast failure rates for ship turbines and pumps. And they can predict more than when equipment will fail, but also the type of failure and why it’s failing.
“We built a solution for them that extended their failure forewarning from four hours to five days. “That’s the kind of impact the tech brings. We don’t know anything about pumps; we’re not domain experts in pumps.”
“So instead of going up and looking through these manuals and figuring out what to do when something breaks and going through that prognostics process, you want the model to basically anticipate what you need and also fetch information that is collected from various different manuals … based on the intelligent understanding of that content and is provided back to the person asking the question.
“The threat surface is so massive that dealing with the types of threats that are emanating
It might also be able to help with the reported pump and turbine problems with the Navy’s Littoral Combat Ships. “That is the exact kind of thing we look into and that we protect on a commercial level with a large number of companies today.”
Then there’s the applicability across all military systems.
An “AI watchman” could prevent ships from colliding with one another since the computers are “constantly looking at sensor data and is making sense of the environment and the situation.”
“There is that safety aspect of using artificial intelligence to augment the level of capability and intelligence available on ships, on tanks, in aircraft, all over, where you almost have an embedded AI technician be part of every military asset.
“That is a capability and it leads to tremendous benefits. And the possibilities may be endless. There’s an easy answer to the question: Where can AI be applied? “It can be applied literally everywhere.
Military Applications of Artificial Intelligence Tech: Questions Remain How Best to Utilise for Operation Scenarios
The Pentagon recently drove tactical trucks with sensors, electronics and other applications powered by commercially-developed artificial intelligence technology as a way to take new steps in more quickly predicting and identifying mechanical failures of great relevance to combat operations.
An Army-industry assessment incorporated attempts to use AI and real-time data analytics for newer, fast-evolving applications of Conditioned Based Maintenance technology.
Advanced computer algorithms, enhanced in some instances through machine learning, enable systems to instantly draw upon vast volumes of historical data as a way to expedite analysis of key mechanical indicators. Real-time analytics, drawing upon documented pools of established data through computer automation, can integrate otherwise disconnected sensors and other on-board vehicle systems.
“We identified some of the challenges in how you integrate sensor data that is delivered from different solutions. You can take unstructured information from maintenance manuals, reports, safety materials, vehicle history information and other vehicle technologies – and use AI to analyze data and draw informed conclusions of great significance to military operators.
Faster diagnostics, of course, enables vehicle operators to anticipate when various failures, such as engine or transmission challenges, may happen in advance of a potentially disruptive battlefield event. Alongside an unmistakable operational benefit, faster Conditioned Based Maintenance activity also greatly streamlines the logistics train, optimizes repairs and reduces costs for the Army.
Army wheeled tactical vehicles, which include things like the Family of Medium Tactical Vehicles and emerging Joint Light Tactical Vehicle, are moving towards using more automation and AI to gather, organise and analyze sensor data and key technical indicators from on-board systems.
“We identified Army data challenges, delivered new sensors – and used different approaches – invariably brings on different ways that data can be delivered to the Army,”
Faster computer processing brings substantial advantages to Army vehicles which increasingly rely upon networked electronics, sensors and C4ISR systems.
“We know there is going to be unmanned systems for the future, and we want to look at unmanned systems and working with teams of manned systems. This involves AI-enabled machine learning in high priority areas we know are going to be long term as well as near term applications.
Technical gains in the area of AI and autonomy are arriving at lightning speed, offering faster, more efficient technical functions across a wide range of platforms. Years ago, the Army began experimenting with “leader-follower” algorithms designed to program an unmanned tactical vehicle to follow a manned vehicle, mirroring its movements.
Autonomous or semi-autonomous navigation brings a range of combat advantages. A truck able to drive itself can, among other things, free up vehicle operators for other high-priority combat tasks. AI enabled condition based maintenance to function through a variety of methods; sensor information can be gathered, organised and then subsequently downloaded, or wirelessly transmitted using cloud technology.
Military is creating robots that can follow orders and understand what they’re told to do—and execute it with minimal supervision.
Military robots have always had limitations like having practically no onboard intelligence and is piloted by remote control. What the military has long wanted instead are intelligent robot teammates that can follow orders without constant supervision.
The robot can take verbal instructions and interpret gestures. But it can also be controlled via a tablet and return data in the form of maps and images so the operator can see exactly what is behind the building, for example.
The team used a hybrid approach to help robots make sense of the world around them. Deep learning is particularly good at image recognition, so algorithms similar to those Google uses to recognize objects in photos let the robots identify buildings, vehicles, and people. As well as identifying whole objects, a robot can recognize key points like the headlights and wheels of a car, helping them work out the car’s exact position and orientation.
Once it has used deep learning to identify an object, the robot uses a knowledge base to pull out more detailed information that helps it carry out its orders. For example, when it identifies an object as a car, it consults a list of facts relating to cars: a car is a vehicle, it has wheels and an engine, and so on. These facts need to be hand-coded and are time consuming to compile, but teams are looking into ways to streamline this like combining deep learning with a knowledge-base-centered approach so a robot can learn and show judgment.
Consider the example of the command “Go behind the farthest truck on the left.” As well as recognizing objects and their locations, the robot has to decipher “behind” and “left,” which depend on where the speaker is standing, facing, and pointing. Its hard-coded knowledge of the environment gives it further conceptual clues as to how to carry out its task.
The robot can also ask questions to deal with ambiguity. If it is told to “go behind the building,” it might come back with: “You mean the building on the right?”
“We have integrated basic forms of all of the pieces needed to enable acting as a teammate. “The robot can make maps, label objects in those maps, interpret and execute simple commands with respect to those objects, and ask for clarification when there is ambiguity in the command.”
Artificial Intelligence Data Platform Identifies Utility for Combat Vehicles
Sensors were installed on Stryker vehicles, and Artificial Intelligence platform ingested and analyzed maintenance manuals and work orders to create a comprehensive maintenance picture.
With that information, the system was able to flag anomalies and predict when components in the vehicles were likely to fail. That information helped the Army more easily spot and track problems in the field and limit the number of breakdowns that took vehicles out of operation.
With the massive volume of information analyzed, the Army could set up maintenance for individual vehicles rather than sending them in groups for scheduled maintenance.
Maintenance data processed by AI-connected vehicles to save manufacture ring time and money because they would know what parts to keep in inventory for likely repairs.
“Wherever a solider is in the world, we will give them information faster and better analyzed. Newer analysis technologies work faster than what humans have time to look at. This is really going to take us to a greater sense of awareness regarding our equipment.”
Primary purpose of platform is to function as a logistics brain for the Army and improve readiness by tracking, analyzing and disseminating Army readiness information through the organisation of logistics data to speed up commander’s decision-making ability,
Overall, the Army system manages large equipment inventories and repairs, such as those needed for combat vehicles, helicopters and trucks along with vast amounts of smaller items spanning from small arms inventories to data services.
“We enable users to go in and out of a system within a very small technical footprint. We make all of that available 24/7 to Army users with system access.”
Enabling greater interoperability, data-sharing and faster network access can shine a spotlight on security issues in a double-pronged manner. In one sense, streamlining information can make data less dispersed and varied, therefore making it easier to protect.
At the same time, some raise the concern that fewer points of access could avail potential intruders an opportunity to launch a more targeted attack and enable them to reach, and potentially cause damage, deeper into networks
Predictive Maintenance and Logistics Use Case Examples of AI Applications Improve Performance
Power of machine learning to detect anomalies. Some existing predictive maintenance systems have analyzed time series data from sensors, such as those monitoring temperature or vibration, in order to detect anomalies or make forecasts on the remaining useful life of components. Sensor data such as anomaly detection on engine vibration data, and images and video of engine condition.
Deep learning’s capacity to analyze very large amounts of high dimensional data can take this to a new level. By layering in additional data, such as audio and image data, from other sensors—including relatively cheap ones such as microphones and cameras—neural networks can enhance and possibly replace more traditional methods.
AI’s ability to predict failures and allow planned interventions can be used to reduce downtime and operating costs while improving production yield.
In some industry examples, using remote on-board diagnostics to anticipate the need for service generate operational value. In a case involving cargo aircraft, AI can extend the life of the plane beyond what is possible using traditional analytic techniques by combining plane model data, maintenance history.
Application of AI techniques such as continuous estimation to logistics can add substantial value across many sectors. through real-time forecasts and behavioral coaching.
AI can optimize routing of delivery traffic, improving fuel efficiency and reducing delivery times.. By using sensors to monitor both vehicle performance and driver behavior, drivers receive real-time coaching, including when to speed up or slow down, optimising fuel consumption and reducing maintenance costs.
One autonomous vehicle uses a morphed tire/track for traction run with a one-handed remote control, non-line of sight manoeuvre with onboard sensors and cameras. Planners are looking to the terrain challenges of dismounted operations to use armed unmanned ground vehicles to provide stand-off force protection.
The concept of self-driving cars has been around for years, but only recently have increasing advances in networking, satellites, and laser equipment made this dream a reality.
Vehicle manufacturers realized that they could use camera systems to relay data to an onboard computer that would process images of the road and create responses. Although we do not have robotic vehicles filling our roadways as of yet, some vehicles already contain numerous autonomous features that make driving easier and safer than ever before. Some models offer assisted parking or braking systems that activate automatically if they sense an issue. Vehicles can sense lane position and make adjustments there as well.
To be useful, robots must navigate the world much as humans do like a test for what could be the future of maintenance work.
The robot isn’t winning any races, but it has endurance. Its battery holds power for three hours and the robot can lower itself onto a charging station when it needs to power up. It’s not light, but could be carried into place on a small vehicle or by a couple of troops. Its limbs can push buttons and push open doors, though it would likely take extra modifications to get it to manipulate doorknobs.
For the tunnel exploration, the robot was lowered into place, and then guided by a joystick. Autonomous movement is possible, but using a remote control allowed the human observers to keep a closer eye on what, exactly, the machine was doing underground.
The robot normally navigates by Light Detection and Ranging, remote sensing technology used to measure distances and 3D mapping of the surrounding environment. To better comprehend the terrain in low-light environments, it is also exploring sensors at the end of its feet, providing a sense of touch. All of this could prove critical to taking place underground tunnels fights.
Finding new ways to incorporate robots and autonomous or semi-autonomous vehicles into warfighting has captured the attention of top commanders but nothing as basic and practical as the gear mule concept has come so close to reality.
1. Importance and necessity of AI transparency is application-specific.
2. Trust must be met across algorithms, data, and outcomes.
3. Users must understand the mechanisms by which systems can be spoofed.
4. Robust and resilient digital capability requires balancing development, operations, and security.
5. Network risk management and network security ownership throughout and across organisations is critical.
6. Applying AI requires a skilled and educated workforce with domain expertise, technical training, and the appropriate tools.
7. Organisations must develop workforce expertise in digital data models
8. Success for users in machine learning requires iteration, experimentation, and learning through early sub-optimal performance.
9. Organisation must build the foundational digital capability to successfully apply AI technologies-- database management, information integration.
10. Gaining competitive advantage through information and analytics is an enterprise-wide endeavor from headquarters to the deployed warfighter.