Bird Gust Soaring Manoeuvre Identification for Energy Efficient Urban Flight #

Freddie Turner, Shane Windsor

Poster session

Abstract #

In addition to riding thermals and updrafts, birds can gain energy by flying through nonuniform wind fields. This has been widely studied in the context of dynamic soaring, where birds such as albatross exploit large-scale wind gradients above the ocean surface to fly long distances with minimal energy expenditure. The same principles of energy extraction can in theory be applied to small scale spatial and temporal gradients (gusts), commonly found in urban environments as wind shear and turbulence. Although birds can be seen to perform complex interactions with the wind, measurement of the local airflow makes it difficult to attribute particular motions with energy gain. This makes it challenging to quantify such ‘gust soaring’ manoeuvres. We avoid issues of airflow measurement by utilising companion data sets of real and simulated urban bird flight trajectories, the former from GPS and IMU ‘backpacks’ carried by lesser black-backed gulls and the latter generated by a wind-aware path planner optimising for energy gain. Analysis of the two data sets using statistical techniques (including machine learning methods) allows us to derive estimates for the prevalence of gust soaring in bird flight and analyse the manoeuvres being performed to gain energy. This research promises to benefit the design of flight control systems for future small uncrewed aerial vehicles, whose range and endurance are restricted by power constraints.