Works of possible interest

General SLAM

• Cadena 2016 – Past, Present, and Future of SLAM
• durrant-whyte 2006 slam tutorial part i

Prerequisites

• g2o paper - graph-based SLAM

Existing SLAM algorithms

• MonoSLAM, works by Andrew Davison focusing on fusion instead of vision-only SLAM

• Maplab (filtering-based) not looking at filtering-based algos

• mentioned in the Chen 2018 Review of VI SLAM paper:

  • ORB-SLAM paper — ORB features
  • VIORB implementation
  • ORB-SLAM3 (improves on ORBSLAM, incl. VI, pub. 2020)
  • Qin 2018 - VINS-Mono: A Robust and Versatile  Monocular Visual-Inertial State Estimator
  • qin 2019 general optimization-based framework (multisensor)

• Deformable SLAM

  • lamarca-2019-defslam
  • badias-2020-morph-dslam

Camera/IMU models

• mkok 2017 using inertial sensors for position and orientation estimation
• Forster 2017 preintegration paper
• Furgale 2013 - calibration of IMU

Kalman filter

• rlabbe kalman/bayesian filters in-python
• Weiss (loose coupling IMU with SLAM using EKF)

Older/classical SLAM papers or related books as recommended by  Cadena 2016

Classical SLAM (1998 - 2004); the first 20 years

• T. Bailey and H. F. Durrant-Whyte, “Simultaneous localisation and mapping (SLAM): Part II,”

•     main probabilistic formulations for SLAM
  • EKF
  • Rao-Blackwellised particle filters
  • MLE
• challenges associated with efficiency and robust data association

• Thrun - Probabilistic robotics (2005)
• Thrun, Stachniss Ch 46 - SLAM “Simultaneous localization and mapping,” in Springer Handbook of Robotics, B. Siciliano and O. 2016, ch. 46, pp. 1153–1176.

Algorithmic analysis age (2004 - 2015)

• Dissanayake G. Dissanayake, S. Huang, Z. Wang, and R. Ranasinghe, “A review of recent developments in simultaneous localization and mapping,” in Proc. Int. Conf. Ind. Inform. Syst., 2011, pp. 477–482.

General Kalman filters:

as recommended by  rlabbe Kalman/Bayesian filters in Python

• Grewal and Andrew’s Kalman Filtering

• Paul Zarchan’s book Fundamentals of Kalman Filtering

• Simo Särkkä

Misc reading

http://link.springer.com/article/10.1007/s40903-015-0032-7 An Overview to Visual Odometry and Visual SLAM: Applications to Mobile Robotics (difference between VIO and VSLAM)

• as far as I understand it, VO doesn’t involve map creation/updates

as recommended by  Cadena 2016

• Lowry 2016 - Visual place recognition: A survey (also reviews topological SLAM)