drones and their applications in robotics and deep learning
A Drone is basically an aircraft or aerial vehicle without a human pilot on board. It is an unmanned vehicle and is commonly referred to as an Unmanned Aerial Vehicle (UAV). Drones are part of an aircraft system which includes an aerial vehicle, a controller from earth and a system to integrate both these operations. The drones may work either under the control of an operator using remote or using computers.
Originally, they were used in military operations. However, they are now widely used in fields such as agricultural, recreational, scientific, commercial, and other applications, such as drone racing, smuggling, infrastructure inspections, aerial photography, product deliveries and policing and surveillance. Many terms are presently used for Drones which basically refer to the same concept.
Components of a Drone
The components of a drone include body, power supply and platform, computing, sensors, actuators, software, loop principles, flight controls and communications.
The software component or the flight stack or the autopilot includes Raspberry Pis, Beagleboards, etc. shielded with NavIO, PXFMini, etc. or designed from scratch such as Nuttx, preemptive-RT Linux, Xenomai, Orocos-Robot Operating System or DDS-ROS 2.0.
Civil use open-source stacks in a Drone include ArduCopter, DroneCode, CrazyFlie, KKMultiCopter, MultiWii, BaseFlight, CleanFlight, BetaFlight, iNav, RaceFlight, OpenPilot, dRonin, LibrePilot, TauLabs and Paparazzi. Drones employ open-loop, closed-loop or hybrid control architectures.
Functions of a Drone
Some of the functions of a Drone are Reactive autonomy by Perceptual control theory, Simultaneous localization and mapping (SLAM), Swarming, future military potential, Cognitive radio and learning capabilities.
Applications of a Drone
The applications of a Drone include recreation, disaster relief, archaeology, conservation of biodiversity and habitat, law enforcement, crime, and terrorism, aerial surveillance, filmmaking, journalism, surveying, cargo transport, mining , Transmission and distribution , Forestry and agriculture, Reconnaissance, attack, scientific research, demining, and target practice.
Drones are precisely used in aerial photography for journalism and film, express shipping and delivery, gathering information or supplying essentials for disaster management, thermal sensor drones for search and rescue operations, geographic mapping of inaccessible terrain and locations, building safety inspections, precision crop monitoring, unmanned cargo transport, law enforcement and border control surveillance and storm tracking and forecasting hurricanes and tornadoes.
Applications of Drones in Robotics: the present and future
Drones as autonomous mobile robots (AMR) are used as autonomous cars, and in cleaning, security, retail hospital and in other applications such as commoditization and modularization of hardware or software.
Drones are used as robots in agriculture. They are used in certain functions including intelligent robotic implements, fresh fruit picking, and in dairy farming. They are also used now as autonomous tractors and exploring the long-term implication of small slow agrobots.
Drones, in the future, will be used as surgical robots, industrial robotic arms, in robotic cleaning and lawn mowing, in warehouses and material handling (autonomous guided vehicles and carts) and in the delivery chain.
Applications of Drones in Deep Learning
Vertical take-off drones are very popular and they perform many tasks using deep learning. Some of the applications of drones in deep learning include large-scale aerial datasets and standardized benchmarks for the training, testing, and evaluation of deep-learning solutions, deep neural networks (DNN) for field aerial robot perception, recurrent networks for state estimation and dynamic identification of aerial vehicles, deep-reinforcement learning for aerial robots in dynamic environments, learning-based aerial manipulation in cluttered environments, decision making or task planning using machine learning for field aerial robots, data analytics and real-time decision making with aerial robots-in-the-loop, aerial robots in agriculture using deep learning, aerial robots in inspection using deep learning, imitation learning for aerial robots and multi aerial-agent coordination using deep learning.