Unmanned Systems Technology Dec/Jan 2020 | Phoenix UAS | Sonar focus | Construction insight | InterGeo 2019 | Supacat ATMP | Adelan fuel cell | Oregon tour | DSEI 2019 | Copperstone Helix | Power management focus

30 December/January 2020 | Unmanned Systems Technology Autonomous flight control system The flight control system was developed by Stirling Dynamics, and is based on a set of flight control laws (FCLs) that calculate the control surface, compressor and valve inputs necessary to achieve a required response, such as a climb or a turn to the left, with the interface to the hardware (actuators and so on) created by the Manufacturing Technology Centre. The flight control hardware selected for use during flight trials was a National instruments Compact RIO, and for the indoor trials a high-powered wi-fi transmitter was used to send data back and forth. The flight control system included the possibility of both manual and automatic modes, the former using a ‘pilot’ to control the aircraft through a radio-control aircraft type control console, the latter being fully autonomous. The FCLs were developed in a simulation environment that included models of the aircraft, environment and actuators. This allowed for full development and testing of the control laws on a desktop PC to give the desired response before going to aircraft trials and testing on the real equipment. The simulation environment consists of three modules – an FD (Flight Dynamics) model, an FCS (Flight Control System) model; and an FCL model. The FD model includes a model of the atmosphere, aerodynamics, and the aircraft dynamics; the FCS model includes all the actuators, compressors, valves and ballonets used to control the aircraft; and the FCL model is the control software that converts pilot/ground station commands, along with feedback from sensors, into demands on the actuation systems. The FCL also consists of two sub-modules – FCC (Flight Control Command) and FCA (Flight Control Algorithm). A simplified schematic of the simulation environment and interfaces between the modules is shown in Fig. 8. The simulation environment was developed in MatLab/Simulink, and the resultant FCL model was used to generate C++ autocode that was loaded onto the onboard flight control computer (note the other parts were for development and testing only). An initial version of the FCL was developed early in the project to help with sizing the flight control surfaces and to predict the likely flight path and response to control inputs. For Phoenix there is an additional set of control parameters compared to a conventional aircraft, which are associated with the variations in buoyancy, namely the control inputs to the compressors and valve feeding and releasing the compressed air into the internal bladder. The time at which the compressors and valve are operated are governed by the control laws based on the aircraft’s velocity, height and proximity to its altitude limits. In conjunction, a set of thresholds was also implemented to trigger the compressors and valve based on the differential pressure across the aircraft’s skin to prevent seams rupturing (at high pressure) or envelope collapse (at low pressure), as shown in Fig. 9 (page 32). Dossier | Phoenix UAS Fig. 7 – Unlike the flight trials, testing the performance of the cells was carried out during daylight Fig. 8 – Diagram of the Phoenix’s flight control system