基于Fisher信息的室内主动SLAM

doi:10.3778/j.issn.1002-8331.2012-0028

摘要/Abstract

摘要： 针对传统的视觉同步定位与地图创建（visual simultaneous localization and mapping，VSLAM）算法在室内弱纹理场景中容易因为特征缺失而定位失败的问题，提出了一种基于最大Fisher信息量云台控制的主动SLAM算法。该方法在经典的ORB-SLAM2框架上进行扩展，增加了Fisher信息场构建模块与云台控制模块。在视觉跟踪的同时，将三维空间划分成若干个体素，根据特征点的空间位置分布更新每个体素的Fisher信息，完成Fisher信息场的构建；当相机获取的图像遇到特征缺失的情况，先找到离相机光心欧式距离最近的体素，以该体素Fisher信息量最大的方向作为相机此时的最优观测方向；计算出坐标变换后相机的偏转角度，通过机载云台实现相机转动到最优观测方向，重新获得场景特征，使得算法在丢失特征之后能够实现自主重定位。将改进后的算法运用到四旋翼无人机仿真平台，结果表明在传统算法失效的情况下，所提算法仍能实时准确地估计无人机位姿，提高了系统的鲁棒性。

关键词: 弱纹理, 位姿估计, Fisher信息

Abstract: Aiming at the problem that traditional visual simultaneous localization and mapping（VSLAM） algorithm is prone to be invalid due to missing features in indoor weak texture scene, an active SLAM algorithm based on maximum Fisher information for tilt control is proposed. The method is extended on the classical ORB-SLAM2 framework, adding Fisher information field construction module and tilt control module. Firstly, the three-dimensional space is divided into several voxels during visual tracking, and the Fisher information of each voxel is updated according to the spatial distribution of features to complete the construction of Fisher information field. Secondly, when the image is acquired by the camera encounters the situation of missing features, the voxel nearest to the optical center of the camera is found, and the direction with the largest Fisher information of this voxel is considered to be the optimal observation direction of the camera. Finally, the deflection angle of the camera after coordinate transformation is calculated, and the camera is rotated to the optimal observation direction through the airborne tilt to obtain the features in scene again, which makes the algorithm realize autonomous relocation after losing features. The improved algorithm is applied to the simulation platform of quadrotor UAV and the results show that the algorithm proposed can still estimate the pose of UAV accurately in real time when the traditional algorithm fails, which improves the robustness of the system.

Key words: weak texture, pose estimation, Fisher information

席永辉, 胡士强. 基于Fisher信息的室内主动SLAM[J]. 计算机工程与应用, 2022, 58(13): 257-264.

XI Yonghui, HU Shiqiang. Indoor Active SLAM Based on Fisher Information[J]. Computer Engineering and Applications, 2022, 58(13): 257-264.

参考文献

[1] BRESSON G，ALSAYED Z，YU L，et al.Simultaneous localization and mapping：A survey of current trends in autonomous driving[J].IEEE Transactions on Intelligent Vehicles，2017，2（3）：194-220.
[2] 王冠政，汪海洋，程志伟，等.vSLAM在无人机平台上的发展研究综述[J].计算机工程与应用，2019，55（14）：8-14.
WANG G Z，WANG H Y，CHENG Z W，et al.Survey of vSLAM based on UAVs[J].Computer Engineering and Applications，2019，55（14）：8-14.
[3] 徐爱功，宋帅，隋心，等.一种单目视觉/INS组合室内定位抗差方法[J].测绘科学，2019，44（12）：1-6.
XU A G，SONG S，SUI X，et al.A single vision/INS combination indoor positioning robust method[J].Science of Surveying and Mapping，2019，44（12）：1-6.
[4] 郭文县，高晨曦，张智，等.基于特征稀疏策略的室内机器人SLAM研究[J].计算机工程与应用，2017，53（16）：110-115.
GUO W X，GAO C X，ZHANG Z，et al.Research on indoor SLAM based on strategy of sparse features[J].Computer Engineering and Applications，2017，53（16）：110-115.
[5] DAVISON A J.Real-time simultaneous localisation and mapping with a single camera[C]//Proceedings of the Ninth IEEE International Conference on Computer Vision，2003：1403-1410.
[6] KLEIN G，MURRAY D.Parallel tracking and mapping for small AR workspaces[C]//Proceedings of the IEEE & ACM International Symposium on Mixed & Augmented Reality，2008：225-234.
[7] MONTEMERLO M，THRUN S，KOLLER D.FastSLAM2.0：An improved particle filtering algorithm for simultaneous localization and mapping that provably converges[C]//Proceedings of the International Joint Conference on Artificial Intelligence，2003：1151-1156.
[8] MUR-ARTAL R，TARDOS J D.ORB-SLAM2：An open-source SLAM system for monocular，stereo and RGB-D cameras[J].IEEE Transactions on Robotics，2017，33（5）： 1255-1262.
[9] 高翔，张涛.视觉SLAM十四讲[M].北京：电子工业出版社，2017：71-74.
GAO X，ZHANG T.14 lectures on visual SLAM[M].Beijing：Electronic Industry Press，2017：71-74.
[10] 李海丰，胡遵河，陈新伟.PLP-SLAM：基于点、线、面特征融合的视觉SLAM方法[J].机器人，2017，39（2）：214-220.
LI H F，HU Z H，CHEN W X.PLP-SLAM：A visual SLAM method based on point-line-plane feature fusion[J].Robot，2017，39（2）：214-220.
[11] ENGEL J，SCH?PS T，CREMERS D.LSD-SLAM： Large?scale direct monocular SLAM[C]//Proceedings of the European Conference on Computer Vision，2014：834-849.
[12] ENGEL J，KOLTUN V，CREMERS D.Direct sparse odometry[J].IEEE Transactions on Pattern Analysis & Machine Intelligence，2018：40（3）：611-625.
[13] LEUTENEGGER S，LYNEN S，BOSSE M，et al.Keyframe-based visual-inertial odometry using nonlinear optimization[J].The International Journal of Robotics Research，2015，34（3）：314-334.
[14] QIN T，LI P L，SHEN S J.VINS-Mono：A robust and versatile monocular visual-inertial state estimator[J].arXiv：1708.03852，2017.
[15] 王延东.多位姿信息融合的双目视觉惯性里程计研究[D].北京：中国科学院大学，2019.
WANG Y D.Study on binocular visual inertial odometry of multi-pose information fusion[D].Beijing：University of Chinese Academy of Sciences，2019.
[16] 杨祥红，王希彬.无人机主动SLAM技术研究[J].系统仿真技术，2015，11（4）：283-287.
YANG X H，WANG X B.Research on active SLAM technology for UAV[J].System Simulation Technology，2015，11（4）：283-287.
[17] ZHANG Z，SCARAMUZZA D.Perception-aware receding horizon navigation for MAVs[C]//Proceedings of International Conference on Robotics and Automation，2018：2534-2541.
[18] MURALI V，SPASOJEVIC I，GUERRA W，et al.Perception-aware trajectory generation for aggressive quadrotor flight using differential flatness[C]//Proceedings of 2019 American Control Conference（ACC），2019：3936-3943.
[19] EFRON B，HINKLEY D V.Assessing the accuracy of the maximum likelihood estimator：Observed versus expected fisher information[J].Biometrika，1978，65（3）：457-483.
[20] CARLONE L，KARAMAN S.Attention and anticipation in fast visual-inertial navigation[C]//Proceedings of International Conference on Robotics and Automation，2017：3886-3893.
[21] PUKELSHEIM F.Optimal design of experiments[M].[S.l.]：Society for Industrial and Applied Mathematics，2006.
[22] RUBLEE E，RABAUD V，KONOLIGE K，et al.ORB：An efficient alternative to SIFT or SURF[C]//Proceedings of International Conference on Computer Vision，2011：2567-2571.
[23] 董小圆，彭珍瑞，殷红，等.传感器优化布置的距离系数-Fisher信息准则[J].智能系统学报，2017，12（1）：32-37.
DONG X Y，PENG Z R，YIN H，et al.Distance coefficient-Fisher information criterion for optimal sensor placement[J].CAAI Transactions on Intelligent Systems，2017，12（1）：32-37.
[24] BOX M J.Bias in nonlinear estimation[J].Journal of the Royal Statistical Society（Series B：Methodological），1971，33（2）：171-201.
[25] ZHANG Z，SCARAMUZZA D.Beyond point clouds：fisher information field for active visual localization[C]//Proceedings of International Conference on Robotics and Automation，2019：5983-5992.
[26] 胡运红，潘美芹.等式约束优化问题SQP算法的超线性收敛充要条件[J].山东科技大学学报（自然科学版），2006（3）：96-99.
HU Y H，PAN M Q.Sufficient and essential conditions of super?linear convergence for SQP algorithm in equality constrained optimization problem[J].Journal of Shandong University of Science and Technology（Natural Science），2006（3）：96-99.