Dynamic environment simulation using indoor drones

点击翻译按钮获取中文摘要

Unmanned Aerial Vehicles (UAVs) are becoming a common data collection and transport tool for many industries, including agriculture, oil and gas, insurance, search and rescue, infrastructure, real estate and the film industry. In each application, UAVs are required to carry and point specific payloads (usually cameras, but sometimes other sensors such as LIDARs or gas sensors) in order to gather the desired data. Relatively simple feedback control has been used to control UAVs, but with the number of use cases and potential payloads growing each year, the robustness of simple feedback controllers with only rudimentary knowledge of the UAV dynamics is in question. MicroPilot needs techniques for identifying specific vehicle dynamics using commercial autopilot hardware and software so that feed-forward control terms can be included in the autopilot controller to better compensate for unknown and un-modelled UAV dynamics. This research project will develop and validate a set of UAV control and estimation algorithms that identify the UAV dynamics and compensate for them in an adaptive sense. Armed with this innovative technology, MicroPilot's autopilot products will be able to adapt to and compensate for a wide variety of UAVs and payloads without requiring re-programming. This advancement will make MicroPilot more competitive in the global UAV market by making their product more applicable to new industries and use cases as they arise.