Visual attention prediction improves performance of autonomous drone racing agents.
Visual attention prediction improves performance of autonomous drone racing agents.
复制标题
DOI:
10.1371/journal.pone.0264471
复制
发表时间:
2022
期刊:
影响因子:
3.7
通讯作者:
Scaramuzza D
中科院分区:
文献类型:
--
作者:
Pfeiffer C;Wengeler S;Loquercio A;Scaramuzza D
Humans race drones faster than neural networks trained for end-to-end autonomous flight. This may be related to the ability of human pilots to select task-relevant visual information effectively. This work investigates whether neural networks capable of imitating human eye gaze behavior and attention can improve neural networks’ performance for the challenging task of vision-based autonomous drone racing. We hypothesize that gaze-based attention prediction can be an efficient mechanism for visual information selection and decision making in a simulator-based drone racing task. We test this hypothesis using eye gaze and flight trajectory data from 18 human drone pilots to train a visual attention prediction model. We then use this visual attention prediction model to train an end-to-end controller for vision-based autonomous drone racing using imitation learning. We compare the drone racing performance of the attention-prediction controller to those using raw image inputs and image-based abstractions (i.e., feature tracks). Comparing success rates for completing a challenging race track by autonomous flight, our results show that the attention-prediction based controller (88% success rate) outperforms the RGB-image (61% success rate) and feature-tracks (55% success rate) controller baselines. Furthermore, visual attention-prediction and feature-track based models showed better generalization performance than image-based models when evaluated on hold-out reference trajectories. Our results demonstrate that human visual attention prediction improves the performance of autonomous vision-based drone racing agents and provides an essential step towards vision-based, fast, and agile autonomous flight that eventually can reach and even exceed human performances.
登录
查看更多内容
影响因子:
25
作者:
Foehn, Philipp;Romero, Angel;Scaramuzza, Davide
通讯作者:
Scaramuzza, Davide
影响因子:
3.9
作者:
Kang, Byeongkeun;Lee, Yeejin
通讯作者:
Lee, Yeejin
影响因子:
4.3
作者:
Li, Shuo;Ozo, Michael M. O., I;de Croon, Guido C. H. E.
通讯作者:
de Croon, Guido C. H. E.
影响因子:
64.8
作者:
LAND, MF;LEE, DN
通讯作者:
LEE, DN
影响因子:
5.2
作者:
Jung, Sunggoo;Hwang, Sunyou;Shim, David Hyunchul
通讯作者:
Shim, David Hyunchul