Jiannan Zhao enrolled as a PhD Scholar at the University of Lincoln in 2016. In 2017-18 he visited Tsinghua University as part of the STEP2DYNA project funded by the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Skolodowska-Curie grant agreement. During this secondment Jiannan developed the first generation of “locust-inspired collision detector for UAV” and demonstrated real flight with the bio-inspired algorithm on embedded system.
Jiannan has just completed his second 12 month secondment at the Guangzhou University in China as part of the ULTRACEPT project funded by the European Union’s Horizon 2020 Research and Innovation Programme under the Marie Skolodowska-Curie grant agreement. He has been involved in Work Package 1 and 4.
The ultimate objective of my PhD research has been to develop an automatic UAV platform with bio-inspired collision avoidance system. The aim of my secondment to Guangzhou University was to realise agile autonomous UAV flight based on LGMD collision detector.
During my secondment I analysed the challenges during 3D movement of the UAV flight and modelled a novel neural network to overcome these challenges.
The existing algorithms were inadequate for flight scenes. To fully achieve flexible automatic flight the algorithms needed to be enhanced to ensure they were robust against dynamic background noise. During my secondment to Guangzhou University I worked on modelling a robust and efficient locust-inspired algorithm for collision detection. Based on distributed presynaptic interconnections, I have developed a novel model appropriate for agile UAV flight, which can easily filter out insignificant visual cues by discriminating the angular velocity of images.
This model is robust for detecting near range emergent collision in dynamic backgrounds as demonstrated in the following video:
In the next phase of my research, the computational algorithm will be transplanted to embedded systems to achieve efficient automatic flight.
During my secondment I successfully submitted a paper to IEEE Transactions on Cybernetics in July 2020, titled ‘Enhancing LGMD’s Looming Selectivity for UAV Agile Flights with Spatial-temporal Distributed Presynaptic Connections’.
I also joined a group of four Tsinghua University robotic students and competed in the first International Competition for Autonomous Running Intelligent Robots in Beijing. We successfully competed against 32 other teams to take first prize. Read more about the competition here.
These Marie Sklodowska-Curie secondments have provided me access to facilities and recording equipment needed for setting up the UAV platform. Moreover, the weekly meetings with other colleagues of the project has broaden my sights and boosted my research skills.