However, there is such a laboratory at the University of Pennsylvania. As early as 2006, UAV related technologies were explored. Until now, their drone algorithms and technologies have been widely used in agriculture, disaster relief, construction, and entertainment. try. Compared to common drones in photography or logistics, the drones they study have more diverse scenarios, more complicated problems to deal with, and more extensive technical areas involved. It can be said that each and every one of their new developments leads the technological change in the entire drone field and is also a lever to promote the development of drones and artificial intelligence.
Recently, "Programmer" magazine interviewed the Dean of the School of Engineering of the University of Pennsylvania, IEEE, ASME Fellow, Academician of the National Academy of Engineering, and Vijay Kumar at the CCF-GAIR Global Artificial Intelligence and Robotics Summit. He talked with us about the latest research results in his laboratory.
In addition to photography, what can drones do?
Vijay Kumar is an Indian robot scientist. He graduated from Indian Institute of Technology Kanpur (IIT) in 1983 and later studied at Ohio State University. He received a doctorate in mechanical engineering in 1987. Vijay is one of the first scientists in the world to study drones. As early as 2006, Vijay and his team successfully developed prototypes of quadcopters for the first time, and actively promoted drones in multiple practical scenarios. The application below. Ten years later, the drone industry has already become hot and the market scale has increased by 10 billion U.S. dollars. It is estimated that by 2020, the drone market will reach 25 billion U.S. dollars. Although the market is getting hotter, there is still a certain distance between academia and industry. This also makes Vijay frankly, he does not care about the drone market, his immediate focus is still on the drone's four core R & D issues: body type, security, intelligence, speed. Although these four points may seem simple, if launched one by one, it will undoubtedly allow us to see a larger picture of drones. Vijay's latest research results are the best interpretation of these four points. These results include the following categories.
Autonomous Micro UAVs, which can use GPS to navigate in complex 3D environments and can be used for search work and Precision Farming. This drone weighs only 20 grams, through the sensor to plan the route and monitor the current environment, sensors include IMU (Inertial Measurement Unit), camera, laser range finder, altimeter and so on.
SWARMS, Scalable sWarms of Autonomous Robots and Mobile Sensors, In the matrix, each aircraft completes its mission in an autonomous (Autonomic) state, and in the absence of a central control system, it responds only to the response of the orbiting aircraft. Make adjustments to achieve the goal of collaboration.
Micro Bio Robots, which are only 10 to 100 microns in diameter, are braked by biosensors and motors. They can be used for drug discovery, minimally invasive treatment, and micro-assembly. At the same time, they can also be applied in the synthesis of biosensors and biological circuits, helping to form a closed loop of "sensing - driving - communication".
Flying Smart Phones, smart phones are the most common high-performance, low-cost processor and sensor integration. Using drones equipped with smart phones, it is equivalent to having an awesome CPU/GPU, two cameras, a piece of Battery, GPS, Wi-Fi, Bluetooth module, IMU, telecommunication data receiver, and corresponding memory. This undoubtedly gives the aircraft better performance.
Through the above studies, drones can be used in many scenarios. For example, drones can be put into buildings as alarms. If there is an intruder, they will immediately report an alarm; they will search for biochemical/gas leaks in buildings or send aircraft into collapsed buildings or nuclear reactors. , to detect radiation intensity; indoor modeling, building real-time maps in unfamiliar space, and even reflecting some of the interior details; handling heavy objects; building construction keels (currently, an algorithm developed by a Vijay student, an aircraft matrix The rectangular keel can be built on its own. Based on the detection of the surrounding aircraft, the aircraft can choose what material to use, where it will be used, and when it will be used.
In addition, Vijay's research aircraft has many practical uses, such as fine farming, the data captured by thermal cameras and ordinary cameras, analyzing the specific needs of each crop, and feeding the results back to farmers, such as Crop A needs. Watering, Crop B requires fertilization, and Crop C requires pest removal. In this way, drones can also calculate the number of fruits per crop, thus helping farmers estimate crop yields and find disease-causing crops.