Navigating assistance system for quadcopter with deep reinforcement learning

Tung Cheng Wu, Shau Yin Tseng, Chin Feng Lai, Chia Yu Ho, Ying Hsun Lai

研究成果: Conference contribution

4 引文 斯高帕斯(Scopus)

摘要

In this paper, we present a deep reinforcement learning method for quadcopter bypassing the obstacle on the flying path. In the past study, algorithm only control the forward direction about quadcopter. In this letter, we use two function to control quadcopter. One is quadcopter navigating function. It is based on calculating coordination point and find the straight path to goal. The other function is collision avoidance function. It is implemented by deep Q-network model. Both two function will output rotating degree, agent will combine both output and turn direct. Besides, deep Q-network can also make quadcopter fly up and down to bypass the obstacle and arrive at goal. Our experimental result shows that collision rate is 14% after 500 flights. Based on this work, we will train more complex sense and transfer model to real quadcopter.

原文English
主出版物標題Proceedings - 2018 1st International Cognitive Cities Conference, IC3 2018
發行者Institute of Electrical and Electronics Engineers Inc.
頁面16-19
頁數4
ISBN(電子)9781538650592
DOIs
出版狀態Published - 2018 十二月 6
事件1st International Cognitive Cities Conference, IC3 2018 - Okinawa, Japan
持續時間: 2018 八月 72018 八月 9

出版系列

名字Proceedings - 2018 1st International Cognitive Cities Conference, IC3 2018

Other

Other1st International Cognitive Cities Conference, IC3 2018
國家/地區Japan
城市Okinawa
期間18-08-0718-08-09

All Science Journal Classification (ASJC) codes

  • 人工智慧
  • 電腦網路與通信
  • 電腦科學應用
  • 教育
  • 城市研究

指紋

深入研究「Navigating assistance system for quadcopter with deep reinforcement learning」主題。共同形成了獨特的指紋。

引用此