“Mavic Air 2”: used drone that recognises objects better (Photo: dji.com)
Chinese researchers promise lower energy consumption and better object recognition
Drones used for the early detection of fires, traffic counting or finding injured people after natural disasters will become even more reliable in the future thanks to so-called “edge computing”, according to researchers at Yunnan University (http://english.ynu.edu.cn/) and the Chinese Academy of Sciences (https://english.cas.cn/). This is because, according to the experts, it will make flight altitude more reliable. According to the experts, this will optimise flight altitude, speed and other parameters and at the same time ensure that the ability to reliably detect searched objects is at the highest level.
Basis actual flight data
With edge computing, data is not first forwarded for central processing on a server, but processed on the spot. “We first present an effective evaluation metric for actual tasks and construct a transparent energy consumption model based on hundreds of actual flight data to formalise the relationship between energy consumption and flight parameters,” said Jiashun Suo, Xingzhou Zhang, Weisong Shi and Wei Zhou.
In their paper, which they present in the “IEEE Internet of Things Journal”, the scientists go on to write: “Then we present a lightweight, energy-efficient priority decision algorithm based on a large amount of actual flight data to help the system decide on flight parameters.”
Training using simulations
The researchers train and evaluate their “E3-UAV” system in a series of simulations. They use it for the DJI (https://www.dji.com) drone “Mavic Air 2”. The system could also be implemented and tested on other UAVs to further evaluate its potential and generalisability, the researchers say. Furthermore, this work could feed into the development of similar object recognition techniques based on edge computing for robotics applications.