Skip to content

Latest commit

 

History

History
252 lines (162 loc) · 8.13 KB

SLAM-Resource.md

File metadata and controls

252 lines (162 loc) · 8.13 KB

Groups on SLAM

Prof. Micheal Kaess is the author of iSAM. His PhD is under supervision of Prof. Frank Dellaert in Georgia Tech, and Post Doc in Prof. John Leonard in MIT. Now he seems focusing on RBGD mapping, and optimization algorithm in SLAM.

Prof. Sanjiv Singh is the supervisor of Ji Zhang who did a great work on 3D lidar SLAM.

People

Prof. Micheal Kaess, Prof. Sanjiv Singh, Dr. Ji Zhang

Research: Prof. Micheal Kaess

  1. Optimization

    • 2006: Square Root SAM: Simultaneous localization and mapping via square root information smoothing
    • 2008: iSAM: Incremental smoothing and mapping
    • 2012: iSAM2: Incremental Smoothing and Mapping Using the Bayes Tree
    • 2014: RISE: An Incremental Trust-Region Method for Robust Online Sparse Least-Squares Estimation
    • 2009: Covariance Recovery from a Square Root Information Matrix for Data Association
  2. RBGD Mapping

    • 2012: Kintinuous: Spatially Extended KinectFusion
    • 2015: Real-time Large Scale Dense RGB-D SLAM with Volumetric Fusion
  3. Cooperative Mapping

    • 2010: Multiple Relative Pose Graphs for Robust Cooperative Mapping

Research: Prof. Sanjiv Singh, Dr. Ji Zhang

  1. 3D Lider SLAM
    • 2014: LOAM: Lidar Odometry and Mapping in Real-time

Prof. Frank Dellaert is one of the earlist researchers on SLAM. Now they mainly focus on developing backend algorithm for mapping or BA, including GTSAM, DDF-SAM, ILBA.

People

Prof. Frank Dellaert

Research

  1. GTSAM: C++ library for SLAM backend, based on iSAM

    • 2006: Square Root SAM: Simultaneous localization and mapping via square root information smoothing
    • 2008: iSAM: Incremental smoothing and mapping
    • 2012: iSAM2: Incremental Smoothing and Mapping Using the Bayes Tree
    • 2014: RISE: An Incremental Trust-Region Method for Robust Online Sparse Least-Squares Estimation
    • 2010-IROS Subgraph-preconditioned conjugate gradients for large scale slam
  2. DDF-SAM: Decentralized Data Fusion, for cooperative SLAM.

    • 2010: DDF-SAM: Fully Distributed SLAM using Constrained Factor Graphs
    • 2012: Fully Distributed Scalable Smoothing and Mapping with Robust Multi-robot Data Association
    • 2013: DDF-SAM 2.0: Consistent Distributed Smoothing and Mapping
  3. iLBA: Incremental Light Bundle Adjustment, open source

    • 2013: Incremental Light Bundle Adjustment for Robotics Navigation

Focus on VINS, Prof. Stergios Roumeliotis is the second author of MSCKF. Cooperate with google, and participate in Project Tango. Focus on cooperative mapping and application on mobile devices.

People

Prof. Stergios Roumeliotis

Research

  1. Cooperative Mapping

    • An offline VI-SLAM algorithm
    • Create optimization with constraint and solve with Lagrangian multipler
    • All features and KFs are variables
    • Speed up by blocking Hessian, efficient when small number of overlapping features observed
    • Both point features, free lines and "Manhatton lines"
    • 2015: Resource-Aware Large-Scale Cooperative 3D Mapping from Multiple Cell Phones
    • 2016: Large-Scale Cooperative 3D Visual-Inertial Mapping in a Manhattan World
  2. Square Root Inverse Sliding Window Filter (SR-ISWF)

    • Compare with MSCKF
    • 2015: A Square Root Inverse Filter for Efficient Vision-aided Inertial Navigation on Mobile Devices
  3. C-KLAM: Constrained Keyframe-Based Localization and Mapping

    • 2013: C-KLAM: Constrained Keyframe Localization and Mapping for Long-Term Navigation
    • 2014: C-KLAM: Constrained Keyframe-Based Localization and Mapping
  4. Incremental Batch Least-Squares (IBLS): SLAM for resource constrained mobile devices (like Tango)

    • 2015: A Square Root Inverse Filter for Efficient Vision-aided Inertial Navigation on Mobile Devices
    • 2014: Efficient Visual-Inertial Navigation using a Rolling-Shutter Camera with Inaccurate Timestamps
  5. Visual-Inertial Navigation System (VINS) using a Rolling-Shutter Camera

Prof. Anastasios Mourikis is the athor of MSCKF.

People

Prof. John Leonard

Research

  1. Visual SLAM
    • 2016 ICRA: High-Performance and Tunable Stereo Reconstruction
    • 2015 RSS: Monocular SLAM-Supported Object Recognition

Seems more focus on different robot systems.

People

Prof. Roland Siegwart

  1. Calibration

    • Kalibr
  2. VIO

    • 2015: Robust Visual Inertial Odometry Using a Direct EKF-Based Approach (ROVIO)
    • 2013: Keyframe-Based Visual-Inertial SLAM Using Nonlinear Optimization (OKVIS)

To check their works from their youtube channel

People

Prof. Marc Pollefeys, Dr. Lionel Heng

Research

  1. Calibration

    • 2015 JFR: Leveraging Image-based Localization for Infrastructure-based Calibration of a Multi-camera Rig
  2. Visual Localization

    • Large scale
    • Place recognition

People

Prof. Davide Scaramuzza

Research

  1. VINS

    • 2016 TRO: On-Manifold Preintegration for Real-Time Visual-Inertial Odometry
    • 2014 ICRA: SVO: Fast Semi-Direct Monocular Visual Odometry
    • 2014 ICRA: REMODE: Probabilistic, Monocular Dense Reconstruction in Real Time
  2. Monocular Dense Reconstruction

  3. UAV

  4. Event-Camera

Prof. Andrew Davison is one of the most important researchers on SLAM in this world. His important works include: MonoSLAM, KinectFusion, DTAM, SLAM++ (Object based SLAM), ...

People

Prof. Andrew Davison

Research

  1. Dense Visual Odometry

  2. Event-Camera

  3. Object SLAM

Dr. Stefan Leutenegger is the author of OKVIS, he received his PhD degree from Autonomous Systems Lab of ETH.

People

Dr. Stefan Leutenegger, Prof. Andrew Davison

Research

  1. VINS

    • OKVIS
    • 2015 IJRR: Keyframe-based visual–inertial odometry using nonlinear optimization
  2. Dense RBGD SLAM

    • 2015: ElasticFusion: Dense SLAM Without A Pose Graph
  3. Event-Camera

People

Prof. Daniel Cremers.

Research

  1. LSD-SLAM

    • One of the most popular visual SLAM algorithm
    • Semi-dense
  2. DSO: direct sparse odometry

People

Prof. Juan D. Tardos, Prof. Javier Civera

Research

  1. ORB-SLAM

    • One of the most popular visual SLAM algorithm
    • Sparse ORB feature
    • ORB based BoW for loop close detection
  2. Visual-Inertial Direct SLAM

KIT

Dr. Andreas Geiger

T. Barfoot – Autonomous Space Robotics Laboratory, University of Toronto

Blogs and Websites

Tutorials

Topics To Learn

Point Clous to Mesh

Poisson reconstruction, VRIP(volumetric range image processing), ...

SunJian's work

Unsupervised Intrinsic Calibration of Depth Sensors via SLAM, 2013 RSS

Visual Inertial Direct SLAM, UniZar & UniPenn, 2016 ICRA

DPPTAM: Dense piecewise planar tracking and mapping from a monocular sequence, 2015 IROS

DSO: Direct sparse odometry, TUM

Tools to Use

Mendenley: literature management

Some Markdown Editor in Ubuntu