Code underlying publication: Reinforcement Learning Compensated Extended Kalman Filter for Attitude Estimation

This collection contains all code to produce the results of "Reinforcement Learning Compensated Extended Kalman Filter for Attitude Estimation," <em style="color:rgb(51, 51, 51);">2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)</em>, Prague, Czech Republic, 2021, pp. 6854-6859, doi: 10.1109/IROS51168.2021.9635963. This paper leverages reinforcement learning to compensate for the classical extended Kalman filter estimation, i.e., to learn the filter gain from the sensormeasurements. The code is written in python. To use the code, the readers could set up the Python environment according to "requirements.txt." For details, please follow "README.md".

本数据集包含复现论文《用于姿态估计的强化学习补偿扩展卡尔曼滤波器》（Reinforcement Learning Compensated Extended Kalman Filter for Attitude Estimation）所需的全部代码。该论文发表于2021年IEEE/RSJ智能机器人与系统国际会议（IROS），举办地为捷克共和国布拉格，2021年，页码范围为6854至6859，DOI编号为10.1109/IROS51168.2021.9635963。本研究利用强化学习（Reinforcement Learning）对经典扩展卡尔曼滤波器（Extended Kalman Filter，EKF）估计进行补偿，即从传感器测量数据中学习滤波器增益。本代码采用Python语言编写，使用者可根据“requirements.txt”配置Python运行环境，详细操作说明请参阅“README.md”文档。