IoT Techniques and Elements for Drone Package Delivery Networks
If more accuracy is required , differential techniques such as Real Time Kinematics ( RTK ) can be employed , where a local receiver ( sometimes attached to a cell tower or drone mailbox ) reads the GNSS position indicated for a known location , and broadcasts corrections to the drone [ 18 ]. The Verizon RTK system is one example that can routinely achieve centimeter-level precision [ 19 ], making it useful for precision landing , where uncorrected GNSS would not be .
Altitude over ground is another important sensor reading for the drone flight control algorithms . Several alternative altitude determination systems are in common use , sometimes in combination , including : GNSS with RTK , RADAR altimeters , LiDAR altimeters , barometric altimeters , ultrasonic range finders , integration of Inertial Measurement Unit ( IMU ) outputs , and down-pointing depth cameras [ 20 ].
The flight control computer uses all these IoT sensor readings to maintain stable flight , and to direct the drone between waypoints . Many drones also have sensors to report battery voltage , current , temperature and charge state , motor performance , and various internal and environmental measurements . Streams of telemetry from these sensors are made available to auxiliary computers and downlinked to ground networks .
5.1.4 AUXILIARY COMPUTERS
Many cargo drones , especially those intended for autonomous operation use a powerful auxiliary computer to supplement the flight control computer . These are typically multicore or GPU processors optimized for special functions , such as analyzing camera feeds , detecting hazards , precision location of landing positions , advanced mission planning , and supporting formation flying . These are typically fully realized edge computers , with operating systems , middleware , protocol stacks , AI inference capabilities and full cybersecurity .
As a concrete example , the auxiliary computer in many autonomous drones run the sense-andavoid algorithms , where a number of attached cameras view the drone ’ s surroundings , and the auxiliary computer continuously analyzes the images to determine the drone ’ s position , altitude , and if any obstacles ( like human piloted aircraft , other drones , birds , vegetation , power lines , etc .) are in the flight path . If hazards are discovered , the auxiliary computer automatically commands the flight control computer to make evasive maneuvers to avoid the hazard , then return to course .
NASA ’ s Independent Configurable Architecture for Reliable Operations of Unmanned Systems ( ICAROUS ) is an excellent example of sense-and-avoid [ 21 ]. Auxiliary computers are also sometimes used for flight and fleet optimization calculations that try to improve the speed , energy efficiency , safety , reliability , range , performance and resource utilization of large-scale delivery networks .
Journal of Innovation 37