Unmanned Aerial Vehicles (aka UAVs) have tremendous potential in the emergence of novel applications in various areas ranging from Medicine, defense, personal and enterprise security or search and rescue applications for several industries.
Lately, there has been heavy investment in the development of UAVs and multi-UAVs systems that can collaborate and complete missions more efficiently and economically. Emerging technologies, such as 4G/5G networks have significant potential on UAVs equipped with cameras, sensors, and GPS receivers in delivering IoT services. UAV is an all encompassing term that is used to refer to all things that fly while being controlled remotely. The market is one of the most quickly growing and innovative sectors of the IoT and presents a huge commercial and strategic opportunity for telecommunication operators and their technology partners.
Everyone is talking about them, but in plain English, what exactly are drones?
A drone is an unmanned aircraft or ship that is guided remotely or autonomously. Here is a multi-copter drone, named so on account of it’s many propellers.
UAV stands for Unmanned Aerial Vehicle, something that can fly without a pilot on-board. Above is a quad-copter UAV, named so on account of it’s 4 propellers.
This paper focusses on the infusion of Artificial Intelligence into the already available drone tech with the purpose of better capabilities that can be leveraged across multiple industrial segments. The case studies that would be covered address a range of solutions that span Search & Rescue, Low Altitude airspace management, Disaster Response and similar.
The Game of Drones
Having understood the basic terminology used, let us get into a bit more depth as to how the drone technology can be infused with some bearing of intelligence. In the context of any device deemed “smart, we essentially refer to its ability of harnessing valuable date, interpreting it and then using the inferences for decision making. The approach, architecture, protocols and systems that go into making the UAV intelligent is what is being referred to as the “Game of Drones”.
Infusion of AI and Drones
The concept of artificial intelligence widely understood by technologists is the ability of a software program to accept a continuous flow of input data and use this data to build a knowledge base that can be accessed when attempting to arrive at a logical conclusion or decision. The emphasis for such a system is the continuous flow of data and here is where the drones can provide the stream of information to be plugged into the AI module. So, how do these systems communicate and interact in order to provide a solution that is fit for purpose.
The answer lies in the realm of the Internet of Things (IOT).
With a gathering of connected devices all around us, and always on, there is a steady stream of status information being transmitted to any listener.
Consider the illustration as seen here. The homes, Ambulance, vehicles, buildings, people are all invisibly connected through the internet. A drone that is in the airspace above can potentially perform one of the following operations:
- Search and Find ( a thing, a person, a situation )
- Passively scan and report back
- Capture images and relay them back
- Follow and track
Now, where does the monitoring element come in?
Reviewing the possible activities that a drone can perform as indicated above, all of it falls under the uber term of monitoring. Easily adapted to reconnaissance, the crucial advantage of drone technology has is that it can cover large expanses of land in a relatively low time and cost. It is quite evident that much of the expense in using a drone is attributed to the skill and cost of the pilot that is controlling it remotely.
Here is where the opportunity for infusing artificial intelligence becomes a possibility. Instead of a person, it would be an AI that controls the drone, receives the data stream and can draw up inferences and make some decisions as well. This can be a 24×7 arrangement since it is a fully automated end to end system. Artificial intelligence and drones are a match made in tech heaven. Pairing the real-time machine learning technology of AI with the exploratory abilities of unmanned drones gives innovators a multitude of opportunities and solutions. Infineon’s OPTIGATM Connect eSIM solutions are a perfect fit and offer all the required capabilities to enable such applications.
In the following sections of this paper, we shall get into the depth of use cases for emergency response teams that deal with real time situations.
Some organizations are developing a platform for the use of ERTs and safety of the public. The AI powered drones are enabled to map areas, track objects and provide analytical feedback in real-time.
Officers receive raw data captured by a set of drones and turn it into actionable insights for police, fire and other emergency teams. A centralized operations centre enables multiple drone users to collaborate, view what each drone sees and even trace a drone’s flight pattern in real-time.
The AI-powered software assists SWAT teams in gathering scene intelligence, assessing damage after hurricanes and tornadoes and even employs thermal imaging to locate missing persons. The OPTIGATM Trust M would help to properly authenticate eligible entities and can help to setup the transport encryption in the TLS protocol, to protect the privacy critical video/audio data. OPTIGA ™ Trust M is a secure root of trust which stores certification and its keys in a tamper resistant manner. Also, the TLS handshake is entirely handled inside this chip.
Another use case that has been of immense interest is the incorporation of neural network’s into the drone abilities of camera based search. A recent article published in one of the leading technology magazines referred to a platform called as Neurala. This platform is a deep learning neural network that helps drones to identify and manage complex situations. A typical situation would be to monitor an area or public space like a stadium or auditorium. Its abilities extend to the inspecting of large industrial equipment, like telephone towers, and generate a real-time damage report allowing the required agencies to arrive on the stop in a timely manner knowing what to expect and what is needed to fix the situation. It has come to be known that a unique application of the Neurala has been done in Africa. The Neurala-powered drones are used to combat elephant poaching in Africa. The artificially intelligent drones use the company’s image recognition technology to monitor elephant herds and spot possible poachers miles before they reach the elephants. Now, isn’t that simply amazing!
With the tremendous demand for online purchases getting higher and higher, technology giants like amazon and Walmart, are looking at highly automated and efficient systems. A start-up in California is building AI powered drones to assist with the mapping and the designing of large-scale warehouses and factories. In addition to tracking the in-house inventory these intelligent devices can perform a depth analysis on current stock via flight video. The secondary benefit of such a system is that it can also use the data from inventory monitoring to develop valuable analytics on popular products hence enabling a faster supply chain. A scan of the shelves could notify a trigger for a reorder and automatically track and add new inventory.
It is very clear that the utilization of drones is increasing in all fields from Agriculture to Industry, from Government to private organizations and from Smart Cities to Rural area monitoring. More than ever before, drones play key problem-solving roles in a variety of sectors — including natural disaster relief, security and construction. With their ability to increase efficiency and improve safety, drones have become important tools for everyone from firefighters to farmers.
With Infineon’s Battery and Wireless Charging Authentication solutions, the efficiency and reliability of drone operations can be greatly improved.
Fig 4: OPTIGA™ TPM (Trusted Platform Module) offers a broad portfolio of standardized security controllers to protect the integrity and authenticity of embedded devices and systems.
Let’s consider the concept of The Internet of Drone Things (IoDT) . It is envisioned as the inevitable direction of Drones backend via Internet of Things, Smart Computer vision, Cloud Computing, advanced wireless communication, big data, and high-end security techniques. With IoDT based implementations, all the existing sectors will become intelligent and smart for performing monitoring, search and rescue and more.
On the Horizon
Smart UAVs are so popular, in fact, that they’re now used on more than 400,000 jobs sites worldwide. And this is only going to grow further.
Imagine a world where the smart drones keep us safe, help to survey the large crop fields to make sure they are healthy, be the first responders in climate triggered disasters and the list goes on.
I would like to leave you with the image of a world that is safer, cleaner and greener.
We as global citizens and women in tech, should endeavor to build and deploy reliable scaled drone systems powered by AI.