ICRA2020论文整理（SLAM + Deep Learning）

参考: https://github.com/PaoPaoRobot/ICRA2020-paper-list

目录: 

1. SLAM

2. Deep Learning in Robotics and Automation

3. Localization

4. Sensor Fusion

5. Calibration and Identification

6. Semantic Scene Understanding

7. Reinforcement Learning for Robotics

8. Omnidirectional Vision

9. Human Detection and Tracking

10. Range Sensing

11. Recognition

12. RGB-D Perception

13. Object Detection, Segmentation and Categorization

14. Autonomous Vehicle Navigation

15. Mapping

---

1. SLAM

• Real-Time Graph-Based SLAM with Occupancy Normal Distributions Transforms

• Spatio-Temporal Non-Rigid Registration of 3D Point Clouds of Plants

• Uncertainty-Based Adaptive Sensor Fusion for Visual-Inertial Odometry under Various Motion Characteristics

• Loam_livox: A Fast, Robust, High-Precision LiDAR Odometry and Mapping Package for LiDARs of Small FoV

• Active SLAM Using 3D Submap Saliency for Underwater Volumetric Exploration

• Are We Ready for Service Robots? the OpenLORIS-Scene Datasets for Lifelong SLAM

• Intensity Scan Context: Coding Intensity and Geometry Relations for Loop Closure Detection

• TextSLAM: Visual SLAM with Planar Text Features

• FlowNorm: A Learning-Based Method for Increasing Convergence Range of Direct Alignment

• Redesigning SLAM for Arbitrary Multi-Camera Systems

• Dynamic SLAM: The Need for Speed

• GradSLAM: Dense SLAM Meets Automatic Differentiation

• Long-Term Place Recognition through Worst-Case Graph Matching to Integrate Landmark Appearances and Spatial Relationships

• Linear RGB-D SLAM for Atlanta World

• Stereo Visual Inertial Odometry with Online Baseline Calibration

• Lidar-Monocular Visual Odometry Using Point and Line Features

• Probabilistic Data Association Via Mixture Models for Robust Semantic SLAM

• Closed-Loop Benchmarking of Stereo Visual-Inertial SLAM Systems: Understanding the Impact of Drift and Latency on Tracking Accuracy

• Metrically-Scaled Monocular SLAM Using Learned Scale Factors

• Inertial-Only Optimization for Visual-Inertial Initialization

• Hierarchical Quadtree Feature Optical Flow Tracking Based Sparse Pose-Graph Visual-Inertial SLAM

• Keypoint Description by Descriptor Fusion Using Autoencoders

• Towards Noise Resilient SLAM

• LAMP: Large-Scale Autonomous Mapping and Positioning for Exploration of Perceptually-Degraded Subterranean Environments

• Modeling Semi-Static Scenes with Persistence Filtering in Visual SLAM

• Broadcast Your Weaknesses: Cooperative Active Pose-Graph SLAM for Multiple Robots

• FlowFusion: Dynamic Dense RGB-D SLAM Based on Optical Flow

• Efficient Algorithms for Maximum Consensus Robust Fitting (I)

• MulRan: Multimodal Range Dataset for Urban Place Recognition

• GPO: Global Plane Optimization for Fast and Accurate Monocular SLAM Initialization

• Large-Scale Volumetric Scene Reconstruction Using LiDAR

• Topological Mapping for Manhattan-Like Repetitive Environments

• Structure-Aware COP-SLAM

• Robust RGB-D Camera Tracking Using Optimal Key-Frame Selection

• Voxgraph: Globally Consistent, Volumetric Mapping Using Signed Distance Function Submaps

• DeepFactors: Real-Time Probabilistic Dense Monocular SLAM

• DOOR-SLAM: Distributed, Online, and Outlier Resilient SLAM for Robotic Teams

• Windowed Bundle Adjustment Framework for Unsupervised Learning of Monocular Depth Estimation with U-Net Extension and Clip Loss

• StructVIO : Visual-Inertial Odometry with Structural Regularity of Man-Made Environments (I)

• Flow-Motion and Depth Network for Monocular Stereo and Beyond

• Online LiDAR-SLAM for Legged Robots with Robust Registration and Deep-Learned Loop Closure

• Hybrid Camera Pose Estimation with Online Partitioning for SLAM

• Analysis of Minima for Geodesic and Chordal Cost for a Minimal 2D Pose-Graph SLAM Problem

• Voxel Map for Visual SLAM

2. Deep Learning in Robotics and Automation

• Learning 3D-Aware Egocentric Spatial-Temporal Interaction Via Graph Convolutional Networks

• C-3PO: Cyclic-Three-Phase Optimization for Human-Robot Motion Retargeting Based on Reinforcement Learning

• AP-MTL: Attention Pruned Multi-Task Learning Model for Real-Time Instrument Detection and Segmentation in Robot-Assisted Surgery

• Automatic Gesture Recognition in Robot-Assisted Surgery with Reinforcement Learning and Tree Search

• Towards Privacy-Preserving Ego-Motion Estimation Using an Extremely Low-Resolution Camera

• ACNN: A Full Resolution DCNN for Medical Image Segmentation

• RoNIN: Robust Neural Inertial Navigation in the Wild: Benchmark, Evaluations, & New Methods

• Segmenting 2K-Videos at 36.5 FPS with 24.3 GFLOPs: Accurate and Lightweight Realtime Semantic Segmentation Network

• Temporally Consistent Horizon Lines

• Full-Scale Continuous Synthetic Sonar Data Generation with Markov Conditional Generative Adversarial Networks

• Real-Time Soft Body 3D Proprioception Via Deep Vision-Based Sensing

• A General Framework for Uncertainty Estimation in Deep Learning

• Learning Local Behavioral Sequences to Better Infer Non-Local Properties in Real Multi-Robot Systems

• Unsupervised Geometry-Aware Deep LiDAR Odometry

• SA-Net: Robust State-Action Recognition for Learning from Observations

• A Generative Approach for Socially Compliant Navigation

• Scalable Multi-Task Imitation Learning with Autonomous Improvement

• Motion2Vec: Semi-Supervised Representation Learning from Surgical Videos

• PointAtrousGraph: Deep Hierarchical Encoder-Decoder with Point Atrous Convolution for Unorganized 3D Points

• Learning Error Models for Graph SLAM

• SMArT: Training Shallow Memory-Aware Transformers for Robotic Explainability

• A 3D-Deep-Learning-Based Augmented Reality Calibration Method for Robotic Environments Using Depth Sensor Data

• Adversarial Feature Training for Generalizable Robotic Visuomotor Control

• Efficient Bimanual Manipulation Using Learned Task Schemas

• BayesOD: A Bayesian Approach for Uncertainty Estimation in Deep Object Detectors

• Learning Object Placements for Relational Instructions by Hallucinating Scene Representations

• FADNet: A Fast and Accurate Network for Disparity Estimation

• Training Adversarial Agents to Exploit Weaknesses in Deep Control Policies

• TRASS: Time Reversal As Self-Supervision

• Federated Imitation Learning: A Novel Framework for Cloud Robotic Systems with Heterogeneous Sensor Data

• Uncertainty Quantification with Statistical Guarantees in End-To-End Autonomous Driving Control

• Autonomously Navigating a Surgical Tool Inside the Eye by Learning from Demonstration

• Learn-To-Recover: Retrofitting UAVs with Reinforcement Learning-Assisted Flight Control under Cyber-Physical Attacks

• Model-Based Reinforcement Learning for Physical Systems without Velocity and Acceleration Measurements

• Towards the Probabilistic Fusion of Learned Priors into Standard Pipelines for 3D Reconstruction

• Learning Natural Locomotion Behaviors for Humanoid Robots Using Human Bias

• Aggressive Online Control of a Quadrotor Via Deep Network Representations of Optimality Principles

• Visual Object Search by Learning Spatial Context

• Salient View Selection for Visual Recognition of Industrial Components

• Low to High Dimensional Modality Reconstruction Using Aggregated Fields of View

• Learning Fast Adaptation with Meta Strategy Optimization

• Deep Neural Network Approach in Robot Tool Dynamics Identification for Bilateral Teleoperation

• Learning Matchable Image Transformations for Long-Term Metric Visual Localization

• OriNet: Robust 3-D Orientation Estimation with a Single Particular IMU

• Learning Densities in Feature Space for Reliable Segmentation of Indoor Scenes

• A Multimodal Target-Source Classifier with Attention Branches to Understand Ambiguous Instructions for Fetching Daily Objects

• On-Board Deep-Learning-Based Unmanned Aerial Vehicle Fault Cause Detection and Identification

• Learning One-Shot Imitation from Humans without Humans

• Adversarial Skill Networks: Unsupervised Robot Skill Learning from Video

• Event-Based Angular Velocity Regression with Spiking Networks

• Visual Odometry Revisited: What Should Be Learnt?

• 3D Scene Geometry-Aware Constraint for Camera Localization with Deep Learning

• ACDER: Augmented Curiosity-Driven Experience Replay

• TrueRMA: Learning Fast and Smooth Robot Trajectories with Recursive Midpoint Adaptations in Cartesian Space

3. Localization

• ROI-Cloud: A Key Region Extraction Method for LiDAR Odometry and Localization

• To Learn or Not to Learn: Visual Localization from Essential Matrices

• Hierarchical Multi-Process Fusion for Visual Place Recognition

• Camera Tracking in Lighting Adaptable Maps of Indoor Environments

• Fast, Compact and Highly Scalable Visual Place Recognition through Sequence-Based Matching of Overloaded Representations

• Vision-Based Multi-MAV Localization with Anonymous Relative Measurements Using Coupled Probabilistic Data Association Filter

• UrbanLoco: A Full Sensor Suite Dataset for Mapping and Localization in Urban Scenes

• Map As the Hidden Sensor: Fast Odometry-Based Global Localization

• Joint Human Pose Estimation and Stereo 3D Localization

• Self-Supervised Deep Pose Corrections for Robust Visual Odometry

• Ultra-High-Accuracy Visual Marker for Indoor Precise Positioning

• Accurate Position Tracking with a Single UWB Anchor

• Association-Free Multilateration Based on Times of Arrival

• Adversarial Feature Disentanglement for Place Recognition across Changing Appearance

• A Fast and Accurate Solution for Pose Estimation from 3D Correspondences

• Ground Texture Based Localization Using Compact Binary Descriptors

• Reliable Data Association for Feature-Based Vehicle Localization Using Geometric Hashing Methods

• Vehicle Localization Based on Visual Lane Marking and Topological Map Matching

• RISE: A Novel Indoor Visual Place Recogniser

• Beyond Photometric Consistency: Gradient-Based Dissimilarity for Improving Visual Odometry and Stereo Matching

• ICS: Incremental Constrained Smoothing for State Estimation

• Drone-Aided Localization in LoRa IoT Networks

• A Fast and Practical Method of Indoor Localization for Resource-Constrained Devices with Limited Sensing

• GN-Net: The Gauss-Newton Loss for Multi-Weather Relocalization

• A Data-Driven Motion Prior for Continuous-Time Trajectory Estimation on SE(3)

• Estimation with Fast Feature Selection in Robot Visual Navigation

• A Tightly Coupled VLC-Inertial Localization System by EKF

• Localization of Inspection Device Along Belt Conveyors with Multiple Branches Using Deep Neural Networks

• Localising PMDs through CNN Based Perception of Urban Streets

• The Complex-Step Derivative Approximation on Matrix Lie Groups

• Hybrid Localization Using Model and Learning-Based Methods: Fusion of Monte Carlo and E2E Localizations Via Importance Sampling

• Measurement Scheduling for Cooperative Localization in Resource-Constrained Conditions

• Quantifying Robot Localization Safety: A New Integrity Monitoring Method for Fixed-Lag Smoothing

• Visual Localization with Google Earth Images for Robust Global Pose Estimation of UAVs

• Relax and Recover: Guaranteed Range-Only Continuous Localization

• SPRINT: Subgraph Place Recognition for Intelligent Transportation

• OneShot Global Localization: Instant LiDAR-Visual Pose Estimation

• Relocalization on Submaps: Multi-Session Mapping for Planetary Rovers Equipped with Stereo Cameras

• DeepTIO: A Deep Thermal-Inertial Odometry with Visual Hallucination

• RSL-Net: Localising in Satellite Images from a Radar on the Ground

• Kidnapped Radar: Topological Radar Localisation Using Rotationally-Invariant Metric Learning

• Global Visual Localization in LiDAR-Maps through Shared 2D-3D Embedding Space

• Unsupervised Learning Methods for Visual Place Recognition in Discretely and Continuously Changing Environments

• LOL: Lidar-Only Odometry and Localization in 3D Point Cloud Maps

• Localising Faster: Efficient and Precise Lidar-Based Robot Localisation in Large-Scale Environments

• Set-Membership State Estimation by Solving Data Association

4. Sensor Fusion

• LiStereo: Generate Dense Depth Maps from LIDAR and Stereo Imagery

• Monocular Visual-Inertial Odometry in Low-Textured Environments with Smooth Gradients: A Fully Dense Direct Filtering Approach

• Gated Recurrent Fusion to Learn Driving Behaviour from Temporal Multimodal Data

• Cooperative Visual-Inertial Odometry: Analysis of Singularities, Degeneracies and Minimal Cases

• A Lightweight and Accurate Localization Algorithm Using Multiple Inertial Measurement Units

• Accelerating the Estimation of Metabolic Cost Using Signal Derivatives: Implications for Optimization and Evaluation of Wearable Robots (I)

• Deep Depth Fusion for Black, Transparent, Reflective and Texture-Less Objects

• LiDAR-Enhanced Structure-From-Motion

• Low Latency and Low-Level Sensor Fusion for Automotive Use-Cases

• Spatiotemporal Camera-LiDAR Calibration: A Targetless and Structureless Approach

• Robot-Assisted and Wearable Sensor-Mediated Autonomous Gait Analysis

• Gaussian Process Preintegration for Inertial-Aided State Estimation

• A Code for Unscented Kalman Filtering on Manifolds (UKF-M)

• Efficient and Precise Sensor Fusion for Non-Linear Systems with Out-Of-Sequence Measurements by Example of Mobile Robotics

• UNO: Uncertainty-Aware Noisy-Or Multimodal Fusion for Unanticipated Input Degradation

• Intermittent GPS-Aided VIO: Online Initialization and Calibration

• A Mathematical Framework for IMU Error Propagation with Applications to Preintegration

• Radar-Inertial Ego-Velocity Estimation for Visually DegradedEnvironments

• Observability Analysis of Flight State Estimation for UAVs and Experimental Validation

• OpenVINS: A Research Platform for Visual-Inertial Estimation

• Decentralized Collaborative State Estimation for Aided Inertial Navigation

• Analytic Combined IMU Integration (ACI^2) for Visual Inertial Navigation

• Second-Order Kinematics for Floating-Base Robots Using the Redundant Acceleration Feedback of an Artificial Sensory Skin

• Clock-Based Time Synchronization for an Event-Based Camera Dataset Acquisition Platform

5. Calibration and Identification

• Unified Intrinsic and Extrinsic Camera and LiDAR Calibration under Uncertainties

• AC/DCC : Accurate Calibration of Dynamic Camera Clusters for Visual SLAM

• Analytic Plane Covariances Construction for Precise Planarity-Based Extrinsic Calibration of Camera and LiDAR

• A Stable Adaptive Observer for Hard-Iron and Soft-Iron Bias Calibration and Compensation for Two-Axis Magnetometers: Theory and Experimental Evaluation

• Extrinsic Calibration of an Eye-In-Hand 2D LiDAR Sensor in Unstructured Environments Using ICP

• Geometric Robot Dynamic Identification: A Convex Programming Approach (I)

• A Novel Calibration Method between a Camera and a 3D LiDAR with Infrared Images

• Online Camera-LiDAR Calibration with Sensor Semantic Information

• Precise 3D Calibration of Wafer Handling Robot by Visual Detection and Tracking of Elliptic-Shape Wafers

• Globally Optimal Relative Pose Estimation for Camera on a Selfie Stick

• Online Calibration of Exterior Orientations of a Vehicle-Mounted Surround-View Camera System

• Learning Camera Miscalibration Detection

6. Semantic Scene Understanding

• Highly Parallelizable Plane Extraction for Organized Point Clouds Using Spherical Convex Hulls
• Boosting Real-Time Driving Scene Parsing with Shared Semantics
• CNN-Based Lidar Point Cloud De-Noising in Adverse Weather
• View-Invariant Loop Closure with Oriented Semantic Landmarks
• Semantic Foreground Inpainting from Weak Supervision
• Fast Panoptic Segmentation Network
• Weakly Supervised Silhouette-Based Semantic Scene Change Detection
• 3DCFS: Fast and Robust Joint 3D Semantic-Instance Segmentation Via Coupled Feature Selection
• Who2com: Collaborative Perception Via Learnable Handshake Communication
• Comparing View-Based and Map-Based Semantic Labelling in Real-Time SLAM
• Generative Modeling of Environments with Scene Grammars and Variational Inference
• SHOP-VRB: A Visual Reasoning Benchmark for Object Perception

7. Reinforcement Learning for Robotics

• Dynamic Actor-Advisor Programming for Scalable Safe Reinforcement Learning

• Discrete Deep Reinforcement Learning for Mapless Navigation

• Learning Multi-Robot Decentralized Macro-Action-Based Policies Via a Centralized Q-Net

• Robust Model-Free Reinforcement Learning with Multi-Objective Bayesian Optimization

• Motor Synergy Development in High-Performing Deep Reinforcement Learning Algorithms

• Barrier-Certified Adaptive Reinforcement Learning with Applications to Brushbot Navigation (I)

• On Simple Reactive Neural Networks for Behaviour-Based Reinforcement Learning

• Predicting Optimal Value Functions by Interpolating Reward Functions in Scalarized Multi-Objective Reinforcement Learning

• Integrated Moment-Based LGMD and Deep Reinforcement Learning for UAV Obstacle Avoidance

• Interactive Reinforcement Learning with Inaccurate Feedback

• Guided Uncertainty-Aware Policy Optimization: Combining Model-Free and Model-Based Strategies for Sample-Efficient Learning

• High-Speed Autonomous Drifting with Deep Reinforcement Learning

8. Omnidirectional Vision

• OmniSLAM: Omnidirectional Localization and Dense Mapping for Wide-Baseline Multi-Camera Systems

• What's in My Room? Object Recognition on Indoor Panoramic Images

• FisheyeDistanceNet: Self-Supervised Scale-Aware Distance Estimation Using Monocular Fisheye Camera for Autonomous Driving

• 360SD-Net: 360° Stereo Depth Estimation with Learnable Cost Volume

• Omnidirectional Depth Extension Networks

• 3D Orientation Estimation and Vanishing Point Extraction from Single Panoramas Using Convolutional Neural Network

9. Human Detection and Tracking

• Natural Scene Facial Expression Recognitionwith Dimension Reduction Network

• Hand Pose Estimation for Hand-Object Interaction Cases Using Augmented Autoencoder

• Accurate Detection and 3D Localization of Humans Using a Novel YOLO-Based RGB-D Fusion Approach and Synthetic Training Data

• Pedestrian Planar LiDAR Pose (PPLP) Network for Oriented Pedestrian Detection Based on Planar LiDAR and Monocular Images

• Wide-Range Load Sensor Using Vacuum Sealed Quartz Crystal Resonator for Simultaneous Biosignals Measurement on Bed

• Joint Pedestrian Detection and Risk-Level Prediction with Motion-Representation-By-Detection

10. Range Sensing

• Super-Pixel Sampler: A Data-Driven Approach for Depth Sampling and Reconstruction
• Physics-Based Simulation of Continuous-Wave LIDAR for Localization, Calibration and Tracking
• A Spatial-Temporal Multiplexing Method for Dense 3D Surface Reconstruction of Moving Objects
• Modeling of Architectural Components for Large-Scale Indoor Spaces from Point Cloud Measurement
• PhaRaO: Direct Radar Odometry Using Phase Correlation
• DeepTemporalSeg: Temporally Consistent Semantic Segmentation of 3D LiDAR Scans

11. Recognition

• Learning Face Recognition Unsupervisedly by Disentanglement and Self-Augmentation
• PARC: A Plan and Activity Recognition Component for Assistive Robots
• Image-Based Place Recognition on Bucolic Environment across Seasons from Semantic Edge Description
• A Multilayer-Multimodal Fusion Architecture for Pattern Recognition of Natural Manipulations in Percutaneous Coronary Interventions
• Action Description from 2D Human Postures in Care Facilities
• CoHOG: A Light-Weight, Compute-Efficient and Training-Free Visual Place Recognition Technique for Changing Environments

12. RGB-D Perception

• ClearGrasp: 3D Shape Estimation of Transparent Objects for Manipulation
• 6D Object Pose Regression Via Supervised Learning on Point Clouds
• YCB-M: A Multi-Camera RGB-D Dataset for Object Recognition and 6DoF Pose Estimation
• Depth Based Semantic Scene Completion with Position Importance Aware Loss
• Self-Supervised 6D Object Pose Estimation for Robot Manipulation
• Panoptic 3D Mapping and Object Pose Estimation Using Adaptively Weighted Semantic Information

13. Object Detection, Segmentation and Categorization

• Stillleben: Realistic Scene Synthesis for Deep Learning in Robotics
• From Planes to Corners: Multi-Purpose Primitive Detection in Unorganized 3D Point Clouds
• Addressing the Sim2Real Gap in Robotic 3D Object Classification
• A Generative Approach towards Improved Robotic Detection of Marine Litter
• Learning to Optimally Segment Point Clouds
• CNN Based Road User Detection Using the 3D Radar Cube
•  
• PST900: RGB-Thermal Calibration, Dataset and Segmentation Network
• Instance Segmentation of LiDAR Point Clouds
• Generation of Object Candidates through Simply Looking Around
• Dilated Point Convolutions: On the Receptive Field Size of Point Convolutions on 3D Point Clouds
• A Water-Obstacle Separation and Refinement Network for Unmanned Surface Vehicles
• Dynamic Anchor Selection for Improving Object Localization
• 3D Object Detection and Tracking Based on Streaming Data
• Object-Centric Stereo Matching for 3D Object Detection
• The Relative Confusion Matrix, a Tool to Assess Classifiablility in Large Scale Picking Applications
• Pose-Guided Auto-Encoder and Feature-Based Refinement for 6-DoF Object Pose Regression
• PrimiTect: Fast Continuous Hough Voting for Primitive Detection
• FarSee-Net: Real-Time Semantic Segmentation by Efficient Multi-Scale Context Aggregation and Feature Space Super-Resolution

14. Autonomous Vehicle Navigation

• Autonomous Navigation in Inclement Weather Based on a Localizing Ground Penetrating Radar
• Robot Navigation in Crowds by Graph Convolutional Networks with Attention Learned from Human Gaze
• Wall Deadlock Evasion Control Based on Rotation Radius Adjustment
• Socially-Aware Reactive Obstacle Avoidance Strategy Based on Limit Cycle
• Multi-Head Attention for Multi-Modal Joint Vehicle Motion Forecasting
• Temporal Information Integration for Video Semantic Segmentation
• Map-Predictive Motion Planning in Unknown Environments
• Using Multiple Short Hops for Multicopter Navigation with Only Inertial Sensors
• An Efficient and Continuous Approach to Information-Theoretic Exploration
• A Feature-Based Underwater Path Planning Approach Using Multiple Perspective Prior Maps
• Automatic LiDAR-Camera Calibration of Extrinsic Parameters Using a Spherical Target

15. Mapping

• A Unified Framework for Piecewise Semantic Reconstruction in Dynamic Scenes Via Exploiting Superpixel Relations

• Keyframe-Based Dense Mapping with the Graph of View-Dependent Local Maps

• Informative Path Planning for Active Mapping under Localization Uncertainty

• Ensemble of Sparse Gaussian Process Experts for Implicit Surface Mapping with Streaming Data

• Robust Method for Removing Dynamic Objects from Point Clouds

• Skeleton-Based Conditionally Independent Gaussian Process Implicit Surfaces for Fusion in Sparse to Dense 3D Reconstruction

• Motion Estimation in Occupancy Grid Maps in Stationary Settings Using Recurrent Neural Networks

• A Divide and Conquer Method for 3D Registration of Inhomogeneous, Partially Overlapping Scans with Fourier Mellin SOFT (FMS)

• Estimating Motion Uncertainty with Bayesian ICP

• Actively Mapping Industrial Structures with Information Gain-Based Planning on a Quadruped Robot

• Efficient Covisibility-Based Image Matching for Large-Scale SfM

• Probabilistic TSDF Fusion Using Bayesian Deep Learning for Dense 3D Reconstruction with a Single RGB Camera

• A Volumetric Albedo Framework for 3D Imaging Sonar Reconstruction

• Map Management Approach for SLAM in Large-Scale Indoor and Outdoor Areas

• A Hierarchical Framework for Collaborative Probabilistic Semantic Mapping

• Autonomous Navigation in Unknown Environments Using Sparse Kernel-Based Occupancy Mapping

• Hybrid Topological and 3D Dense Mapping through Autonomous Exploration for Large Indoor Environments

• Resolving Marker Pose Ambiguity by Robust Rotation Averaging with Clique Constraints