Autonomous navigation of stratospheric balloons using reinforcement learning

By Marc G. Bellemare, Salvatore Candido, Pablo Samuel Castro, Jun Gong, Marlos C. Machado, Subhodeep Moitra, Sameera S. Ponda, Ziyun Wang. 2020

408 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Atmospheric Science, Management Science and Operations Research, Reinforcement Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Atmospheric Science

In Management Science and Operations Research

In Reinforcement Learning

Citations in the corpus

Break-It-Fix-It: Unsupervised Learning for Program Repair (2021, background): “, 2020), as well as from synthetic data to real data (including sim2real) (Wang et al., 2015; Ganin & Lempitsky, 2015; Richter et al., 2016; Peng et al., 2018; Hellendoorn et al., 2019; Xu et al., 2020; Bellemare et al., 2020), as synthetic data can be easier to obtain than real data.”
Same State, Different Task: Continual Reinforcement Learning without Interference (2021, background): “2018), and even real-world settings (Bellemare et al. 2020).”
Acme: A Research Framework for Distributed Reinforcement Learning (2020, background): “, 2020), stratospheric balloon navigation (Bellemare et al., 2020), nuclear fusion (Degrave et al.”
WILDS: A Benchmark of in-the-Wild Distribution Shifts (2020, neutral): “However, due to privacy and other reasons, these datasets are not publicly available. realistic in contexts where real data is much harder to acquire than synthetic data (Bellemare et al., 2020).”
Reaching the limit in autonomous racing: Optimal control versus reinforcement learning (2023, background): “Control systems are at the core of every real-world robot and are deployed in an ever-increasing number of applications, such as self-driving cars (1, 2), search and rescue (3), flying cars (4), autonomous vehicle racing (5, 6), navigation of stratospheric balloons (7), and extraterrestrial explorat”
Vygotskian Autotelic Artificial Intelligence: Language and Culture Internalization for Human-Like AI (2022, methodology): “…methods (deep rl ) have recently made the headlines by solving a wealth of problems we thought only humans could solve: playing Atari video games at super-human level, 2 beating chess and go world champions, 3 controlling stratospheric balloons 4 or even maintaining plasma in fusion reactors.”
Flow: A Modular Learning Framework for Mixed Autonomy Traffic (2017, neutral): “…methodology for sequential decision making [78], [79], which inherits from machine learning and optimal control, and has demonstrated success in complex, data-rich problems such as Atari games [80], 3D locomotion and manipulation [81], [82], [83], and navigation of stratospheric balloons [84].”
Deep Reinforcement Learning at the Edge of the Statistical Precipice (2021, methodology): “[7] Marc G Bellemare, Salvatore Candido, Pablo Samuel Castro, Jun Gong, Marlos C Machado, Subhodeep Moitra, Sameera S Ponda, and Ziyu Wang.”
Magnetic control of tokamak plasmas through deep reinforcement learning (2022, methodology): “The RL approach, already used successfully in a number of challenging applications in other domains [11–13], enables intuitive setting of performance objectives, shifting the focus towards what should be achieved, rather than how.”