ELC + ABS

The growth in capability of computing power and sensors available on small and cheap open source hardware platforms has enabled the application of increasingly sophisticated data fusion techniques to solve the problem of providing robust estimates of flight vehicle states (orientation, velocity and position) under a wide range of operating conditions. In this presentation Paul Riseborough presents recent developments in application of data fusion techniques to the ArduPilot and PX4 codebase which enables data from gyroscopes, accelerometers, magnetometers, range finders, optical flow, GPS and barometric pressure to be combined to estimate vehicle states. He will discuss some of the design challenges associated with development of software used across a wide range of mission and vehicle types, and the data fusion techniques that can adopted in the future as processing capacity grows.