The small Unmanned Aerial System (sUAS) market continues to develop at an impressive pace and is unmarred by the challenges of COVID-19. Global tech market advisory firm, ABI Research, predicts strong growth of the industry worth $92 billion by 2030.

While the pandemic has dented consumer shipments and hindered commercial rollout, this has been mitigated by increased use of drones for public service responses and surveillance by both local and national police forces.

Looking forward, new regulatory changes and the slow rollout of remote ID and 5G will enable an enormous upscaling of drone operations, from single remotely operated aircraft to semi-autonomous fleets that will be able to operate beyond visual line of sight courtesy of impending unmanned traffic management (UTM) infrastructure. This will provide the base from which companies like Amazon can launch drone delivery services.

“We have gone through various phases of the drone industry, from its genesis in the military to the proliferation of consumer drones. Since Chinese developer DJI monopolized that space, the attention has shifted to commercial applications,” said Rian Whitton, senior robotics analyst at ABI Research. “While some of the initial hype has subsided, providers and end-users are refocusing on developing the necessary supporting infrastructure and services to make drone technology viable at scale.”

Drone pilots

Overall, the drone market is set to be worth $92 billion by 2030, with a CAGR rate of 25% over the $9.5 billion in annual revenue for 2020. Of this revenue, 70% is in the commercial sector ($63 billion).

The largest number of drone registrations are currently in the United States, where the Federal Aviation Authority (FAA) tracks 1.7 million consumer drone pilots and 400,000 commercial operators. China is catching up with 400,000 registered drones, while the European Union (EU) has over one million registrants. Among the biggest markets are security and industrial inspection, with growing opportunities in delivery, agriculture, and emergency services.

With their involvement in the public response to the pandemic, drone companies highlighted their value. Now, the story of the next decade will be the development of key technologies like edge computing, cloud services, and 5G connectivity enabling mass deployments, in tandem with regulatory harmonization.

“While cloud services will help enable the collection and orchestration of massive amounts of data, 5G will significantly reduce latency for mission-critical drone operations. The advancement of edge computing and processing hardware will also be important, as drones can be untethered from the remote operation and become truly autonomous,” Whitton said.

The post ABI Research forecasts ‘strong growth’ in drone industry post-COVID-19 appeared first on Artificial Intelligence.

Army researchers developed a reinforcement learning approach that allows swarms of unmanned aerial and ground vehicles to optimally accomplish various missions while minimizing performance uncertainty.

Swarming is a method of operations where multiple autonomous systems act as a cohesive unit by actively coordinating their actions.

Army researchers said future multi-domain battles will require swarms of dynamically coupled, coordinated heterogeneous mobile platforms to overmatch enemy capabilities and threats targeting U.S. forces.

The Army is looking to swarming technology to be able to execute time-consuming or dangerous tasks, said Dr. Jemin George of the U.S. Army Combat Capabilities Development Command’s Army Research Laboratory.

“Finding optimal guidance policies for these swarming vehicles in real-time is a key requirement for enhancing warfighters’ tactical situational awareness, allowing the U.S. Army to dominate in a contested environment,” George said.

Reinforcement learning provides a way to optimally control uncertain agents to achieve multi-objective goals when the precise model for the agent is unavailable; however, the existing reinforcement learning schemes can only be applied in a centralized manner, which requires pooling the state information of the entire swarm at a central learner. This drastically increases the computational complexity and communication requirements, resulting in unreasonable learning time, George said.

In order to solve this issue, in collaboration with Prof. Aranya Chakrabortty from North Carolina State University and Prof. He Bai from Oklahoma State University, George created a research effort to tackle the large-scale, multi-agent reinforcement learning problem. The Army funded this effort through the Director’s Research Award for External Collaborative Initiative, a laboratory program to stimulate and support new and innovative research in collaboration with external partners.

The main goal of this effort is to develop a theoretical foundation for data-driven optimal control for large-scale swarm networks, where control actions will be taken based on low-dimensional measurement data instead of dynamic models.

The current approach is called Hierarchical Reinforcement Learning, or HRL, and it decomposes the global control objective into multiple hierarchies – namely, multiple small group-level microscopic control, and a broad swarm-level macroscopic control.

“Each hierarchy has its own learning loop with respective local and global reward functions,” George said. “We were able to significantly reduce the learning time by running these learning loops in parallel.”

According to George, online reinforcement learning control of swarm boils down to solving a large-scale algebraic matrix Riccati equation using system, or swarm, input-output data.

The researchers’ initial approach for solving this large-scale matrix Riccati equation was to divide the swarm into multiple smaller groups and implement group-level local reinforcement learning in parallel while executing a global reinforcement learning on a smaller dimensional compressed state from each group.

Their current Hierarchical Reinforcement Learning scheme uses a decupling mechanism that allows the team to hierarchically approximate a solution to the large-scale matrix equation by first solving the local reinforcement learning problem and then synthesizing the global control from local controllers (by solving a least squares problem) instead of running a global reinforcement learning on the aggregated state. This further reduces the learning time.

Experiments have shown that compared to a centralized approach, HRL was able to reduce the learning time by 80% while limiting the optimality loss to 5%.

“Our current HRL efforts will allow us to develop control policies for swarms of unmanned aerial and ground vehicles so that they can optimally accomplish different mission sets even though the individual dynamics for the swarming agents are unknown,” George said.

George stated that he is confident that this research will be impactful on the future battlefield, and has been made possible by the innovative collaboration that has taken place.

“The core purpose of the ARL science and technology community is to create and exploit scientific knowledge for transformational overmatch,” George said. “By engaging external research through ECI and other cooperative mechanisms, we hope to conduct disruptive foundational research that will lead to Army modernization while serving as Army’s primary collaborative link to the world-wide scientific community.”

The team is currently working to further improve their HRL control scheme by considering optimal grouping of agents in the swarm to minimize computation and communication complexity while limiting the optimality gap.

They are also investigating the use of deep recurrent neural networks to learn and predict the best grouping patterns and the application of developed techniques for optimal coordination of autonomous air and ground vehicles in Multi-Domain Operations in dense urban terrain.

The post Hierarchical Reinforcement Learning helping Army advance drone swarms appeared first on Artificial Intelligence.

Cleared contractors are found in massive, hyper-modern companies situated in manicured industrial parks. They’re also found in small businesses that have been in existence since WWII. No matter where they are or how big their operating base, they will be affected by Artificial Intelligence (AI). What began as an almost magical way to collate data has turned into a means of analyzing and even predicting based upon the collected information. What began as a good way to streamline boring employee work became in some companies a robot which will do the work. All of this comes with a price. For the secret, classified world, it can be monumental.

HOW AI IS RESHAPING THE DRONE INDUSTRY

I noticed one colleague who cherished his corner office. He identified one of his own weaknesses there, however. He knew that if he faced the window, he’d tend to daydream. So, he turned himself around and placed his workstation facing away from the window in order to better concentrate. I pointed out that anyone with a seeing eye could observe his classified computer from outside. “How?” he demanded. “We’re three floors up!” I gave him a company brochure which showed some of the capabilities of CCTV cameras today. “And those are the good guys watching!” I commented. “What about those who secretly observe what you are writing and reading?” He turned his computer around so its back was to the window again. I hope he doesn’t daydream too much.

A simple introduction to how cameras, computers, and tireless eyes can spy on us is to ask yourself this question. How often am I recorded during the day? Who sees me?  Who hears me?  What can my company legally do to watch or monitor my every move, who I talk to, and when? Remember all the old movies of masses of industrial workers clocking in and out at the factory? You do this today whenever you swipe into a cleared area. What and who authorizes the use of the data collected? What can your boss read of your emails? What can he do to monitor what you do during the day? What, if anything, of your work life is private? If you have a security clearance, the expectation should be that none of your work-day activities are outside of the oversight and careful eye of government monitoring.

ACCESS CONTROL AND YOUR PERSONAL IDENTITY

As citizens we are faced with great challenges from AI, but as clearance holders we can only anticipate some of the many issues to face us. As access control becomes greater, under what circumstances are we compelled to agree to our identity being shared? Consider that now physical facial identification is the target means of future access control. Your face becomes your access “badge”. In some plants, fingerprint codes already are. So ask yourself, what level of protection is necessary? What happens to this data of your many facial identifying data points? This is a real concern.

Studies in China have shown that a man, once facially identified, cannot escape from any major city. Even the smallest photographic capture by a poor CCTV can locate him once his facial ID is entered. For police and counter-terrorist investigations, this is a godsend. For free people, it is worrisome. Consider that the data about you, stored by your company, could be given to researchers who can find you wherever you go. The issues for someone committing a crime are obvious. But what if you are a husband cheating on your wife? Is this something a company can give to a private investigator? Could a pre-employment screening suddenly become not just about the information that’s included in a standard background check, but what is available through other online monitoring systems?

This only touches a single aspect of AI’s new field of research. China’s vast investment in machine learning devotes the largest proportion of its money to the study of computer vision. The Russians have deduced that collecting data by AI saves them vast amounts of human investment in their disinformation projects to influence foreign governments. What might seem to a company as a research question by a legitimate firm seeking data for some honest research could be manipulated by adversaries for use in attacking our democracy. As companies continue to seek out more and more data, it’s critical that the methods to protect it are equally robust.

The post How Artificial Intelligence Creates New Security Risk appeared first on Artificial Intelligence.