• Stars
    star
    226
  • Rank 176,514 (Top 4 %)
  • Language
    C++
  • License
    BSD 3-Clause "New...
  • Created about 3 years ago
  • Updated about 1 year ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Fast, efficient and accurate multi-resolution, multi-sensor 3D occupancy mapping

Wavemap

3D reconstruction of Newer College's Cloister

Hierarchical, multi-resolution volumetric mapping

Wavemap achieves state-of-the-art memory and computational efficiency by combining Haar wavelet compression and a coarse-to-fine measurement integration scheme. Advanced measurement models allow it to attain exceptionally high recall rates on challenging obstacles like thin objects.

The framework is very flexible and supports several data structures, measurement integration methods, and sensor models out of the box. The ROS interface can, for example, easily be configured to fuse multiple sensor inputs, such as a LiDAR configured with a range of 20m and several depth cameras up to a resolution of 1cm, into a single map.

Documentation

The framework's documentation is hosted on GitHub Pages.

Table of contents

Paper

A technical introduction to the theory behind the wavemap is provided in our open-access RSS paper.

Abstract
Volumetric maps are widely used in robotics due to their desirable properties in applications such as path planning, exploration, and manipulation. Constant advances in mapping technologies are needed to keep up with the improvements in sensor technology, generating increasingly vast amounts of precise measurements. Handling this data in a computationally and memory-efficient manner is paramount to representing the environment at the desired scales and resolutions. In this work, we express the desirable properties of a volumetric mapping framework through the lens of multi-resolution analysis. This shows that wavelets are a natural foundation for hierarchical and multi-resolution volumetric mapping. Based on this insight we design an efficient mapping system that uses wavelet decomposition. The efficiency of the system enables the use of uncertainty-aware sensor models, improving the quality of the maps. Experiments on both synthetic and real-world data provide mapping accuracy and runtime performance comparisons with state-of-the-art methods on both RGB-D and 3D LiDAR data. The framework is open-sourced to allow the robotics community at large to explore this approach.

Please cite it when using wavemap for research:

Reijgwart, V., Cadena, C., Siegwart, R., & Ott, L. (2023). Efficient volumetric mapping of multi-scale environments using wavelet-based compression. Proceedings of Robotics: Science and System XIX.

@INPROCEEDINGS{reijgwart2023wavemap,
    author = {Reijgwart, Victor and Cadena, Cesar and Siegwart, Roland and Ott, Lionel},
    journal = {Robotics: Science and Systems. Online Proceedings},
    title = {Efficient volumetric mapping of multi-scale environments using wavelet-based compression},
    year = {2023-07},
}

Note that the code has significantly improved since the paper was written. Wavemap is now up to 10x faster, thanks to new multi-threaded measurement integrators, and uses up to 50% less RAM, by virtue of new memory efficient data structures inspired by OpenVDB.

More Repositories

1

kalibr

The Kalibr visual-inertial calibration toolbox
C++
4,357
star
2

maplab

A Modular and Multi-Modal Mapping Framework
C++
2,610
star
3

voxblox

A library for flexible voxel-based mapping, mainly focusing on truncated and Euclidean signed distance fields.
C++
1,336
star
4

rotors_simulator

RotorS is a UAV gazebo simulator
C++
1,245
star
5

okvis

OKVIS: Open Keyframe-based Visual-Inertial SLAM.
C++
1,158
star
6

rovio

C++
1,126
star
7

segmap

A map representation based on 3D segments
C++
1,070
star
8

ethzasl_msf

MSF - Modular framework for multi sensor fusion based on an Extended Kalman Filter (EKF)
C++
985
star
9

lidar_align

A simple method for finding the extrinsic calibration between a 3D lidar and a 6-dof pose sensor
C++
836
star
10

hfnet

From Coarse to Fine: Robust Hierarchical Localization at Large Scale with HF-Net (https://arxiv.org/abs/1812.03506)
Python
739
star
11

mav_active_3d_planning

Modular framework for online informative path planning.
C++
564
star
12

mav_trajectory_generation

Polynomial trajectory generation and optimization, especially for rotary-wing MAVs.
C++
548
star
13

polygon_coverage_planning

Coverage planning in general polygons with holes.
C++
528
star
14

aerial_mapper

Real-time Dense Point Cloud, Digital Surface Map (DSM) and (Ortho-)Mosaic Generation for UAVs
C++
524
star
15

voxgraph

Voxblox-based Pose graph optimization
C++
513
star
16

robust_point_cloud_registration

Robust Point Cloud Registration Using Iterative Probabilistic Data Associations ("Robust ICP")
C++
513
star
17

mav_voxblox_planning

MAV planning tools using voxblox as the map representation.
Makefile
463
star
18

hand_eye_calibration

Python tools to perform time-synchronization and hand-eye calibration.
Python
439
star
19

dynablox

Real-time detection of diverse dynamic objects in complex environments.
C++
436
star
20

voxblox-plusplus

A volumetric object-level semantic mapping framework.
C++
409
star
21

mav_control_rw

Control strategies for rotary wing Micro Aerial Vehicles using ROS
C
350
star
22

ethzasl_sensor_fusion

time delay single and multi sensor fusion framework based on an EKF
C++
327
star
23

nbvplanner

A real-time capable exploration and inspection path planner (next best view planning)
C++
295
star
24

panoptic_mapping

A flexible submap-based framework towards spatio-temporally consistent volumetric mapping and scene understanding.
C++
275
star
25

ethzasl_icp_mapping

3D mapping tools for robotic applications
C++
268
star
26

okvis_ros

OKVIS: Open Keyframe-based Visual-Inertial SLAM (ROS Version)
C++
256
star
27

versavis

An Open Versatile Multi-Camera Visual-Inertial Sensor Suite
C++
256
star
28

kitti_to_rosbag

Dataset tools for working with the KITTI dataset raw data ( http://www.cvlibs.net/datasets/kitti/raw_data.php ) and converting it to a ROS bag. Also allows a library for direct access to poses, velodyne scans, and images.
C++
248
star
29

laser_slam

This package provides an end-to-end system to laser-based graph SLAM using laser point clouds.
C++
248
star
30

geodetic_utils

Simple library for converting coordinates to/from several geodetic frames (lat/lon, ECEF, ENU, NED, etc.)
C++
247
star
31

COIN-LIO

πŸͺ™ COIN-LIO: Complementary Intensity-Augmented LiDAR Inertial Odometry (ICRA 2024)
C++
245
star
32

image_undistort

A compact package for undistorting images directly from kalibr calibration files. Can also perform dense stereo estimation
C++
245
star
33

ethzasl_ptam

Modified version of Parallel Tracking and Mapping (PTAM)
C++
235
star
34

cblox

Voxblox-based submapping
C++
207
star
35

aslam_cv2

C++
196
star
36

glocal_exploration

Efficient local and global exploration on submap collections with changing past pose estimates.
C++
186
star
37

volumetric_mapping

A repository for 3D volumetric (occupancy) maps, providing a generic interface for disparity map and pointcloud insertion, and support for custom sensor error models.
C++
186
star
38

vgn

Real-time 6 DOF grasp detection in clutter.
Python
181
star
39

hierarchical_loc

Deep image retrieval for efficient 6-DoF localization
Python
172
star
40

orb_slam_2_ros

ROS interface for ORBSLAM2!!
C++
171
star
41

mav_dji_ros_interface

Interface of DJI autopilot based on its OSDK (3.2)
C++
156
star
42

lidar_undistortion

Catkin package that provides lidar motion undistortion based on an external 6DoF pose estimation input.
C++
145
star
43

programming_guidelines

This repository contains style-guides, discussions, eclipse/emacs auto-formatter for commonly used programming languages
Emacs Lisp
139
star
44

tsdf-plusplus

TSDF++: A Multi-Object Formulation for Dynamic Object Tracking and Reconstruction
C++
135
star
45

odom_predictor

Integrates an IMU to predict future odometry readings
C++
134
star
46

depth_segmentation

A collection of segmentation methods working on depth images
C++
133
star
47

grid_map_geo

Geolocalization for grid map using GDAL.
C++
129
star
48

neuralblox

Real-time Neural Representation Fusion for Robust Volumetric Mapping
Python
127
star
49

StructuralInspectionPlanner

ASL Structural Inspection Planner
C++
108
star
50

phaser

A robust pointcloud registration pipeline based on correlation.
C++
106
star
51

eth_supermegabot

Instructions for ETH center for robotics summer school 2019.
Python
102
star
52

terrain-navigation

Repository for Safe Low Altitude Navigation in steep terrain for fixed-wing Aerial Vehicles
C++
98
star
53

waypoint_navigator

Stand-alone waypoint navigator
C++
96
star
54

ethzasl_xsens_driver

Driver for xsens IMUs
Python
96
star
55

mav_tools_public

General launch files, parameters and wiki entries on our systems and related issues
95
star
56

data-driven-dynamics

Data Driven Dynamics Modeling for Aerial Vehicles
Python
94
star
57

reinmav-gym

Reinforcement Learning framework for MAVs using the OpenAI Gym environment
Python
93
star
58

cuckoo_time_translator

algorithms for synchronizing clocks
C++
88
star
59

minkindr

A minimal library for transformations, following the kindr interface. Uses active quaternions of rotation in Hamilton notation.
C++
88
star
60

unreal_airsim

Simulation interface to Unreal Engine 4 based on the AirSim plugin.
C++
87
star
61

waverider

RMPs on multi-resolution occupancy maps for efficient reactive collision avoidance
87
star
62

ethz_piksi_ros

ROS drivers for the Piksi RTK GPS module
C++
85
star
63

sl_sensor

SL Sensor: An open-source, real-time and ROS-based structured light sensor for high accuracy construction robotic applications
C++
84
star
64

voxblox_ground_truth

Create ground truth voxblox maps from Gazebo worlds or .ply files
C++
83
star
65

vicon_bridge

This is a driver providing data from VICON motion capture systems. It is based on the vicon_mocap package from the starmac stacks. Additionally, it can handle multiple subjects / segments and allows to calibrate an origin of the vehicle(s) as this is somehow tedious with the VICON Tracker.
C++
80
star
66

ros-system-monitor

System monitoring tools for ROS.
Python
80
star
67

navrep

Python
73
star
68

curves

A library of curves for estimation.
C++
72
star
69

schweizer_messer

Programming tools for robotics.
C++
65
star
70

time_autosync

Automatically syncs a camera to a rigidly attached IMUs time frame
C++
63
star
71

unreal_cv_ros

Unreal CV ROS Perception Simulator
Python
62
star
72

ai_for_robotics

Programming Exercises Accompanying the Lecture "Artificial Intelligence for Robotics"
Python
60
star
73

lcd

Line Clustering and Description for Place Recognition
C++
59
star
74

trajectory_toolkit

Python tool for analyzing and evaluating trajectory data
Python
59
star
75

dataset_tools

Loader for the generic ASL dataset formats.
MATLAB
58
star
76

rl-navigation

OpenEdge ABL
57
star
77

asl-student-templates

Templates and overview information for student projects at ASL
PostScript
56
star
78

libseekthermal

Driver library for Seek Thermal imaging devices
C++
55
star
79

reactive_avoidance

Reactive obstacle avoidance using raytracing or lidars
C++
52
star
80

plotty

matplotlib-cpp with Eigen interfaces.
C++
52
star
81

forest_gen

Generates randomized Poisson forests to use for UAV collision avoidance evaluations.
Python
49
star
82

3d_vsg

3D Variable Scene Graphs for long-term semantic scene change prediction.
Python
49
star
83

sampling_based_control

Jupyter Notebook
47
star
84

mav_comm

This repository contains message and service definitions used for mavs. All future message definitions go in here, existing ones in other stacks should be moved here where possible.
C++
46
star
85

tmplanner

Terrain monitoring planner
C++
45
star
86

3d3l

Deep Learned Keypoint Detection and Description for 3D LiDARs
Python
44
star
87

fgsp

Jupyter Notebook
44
star
88

autolabel

A project for computing high-quality ground truth training examples for RGB-D data.
Python
43
star
89

mav_gtsam_estimator

A GTSAM based state estimation framework.
C++
43
star
90

visensor_node

Visual inertial SLAM sensor ROS node.
C++
43
star
91

Learn-to-Calibrate

We utilize deep reinforcement learning to obtain favorable trajectories for visual-inertial system calibration.
C++
43
star
92

cvae_exploration_planning

Learning informed sampling distributions and information gains for efficient exploration planning.
Python
42
star
93

active_grasp

Closed-loop next-best view planning for grasp detection in clutter.
Python
41
star
94

two_state_information_filter

C++
41
star
95

ssc_exploration

Incremental 3D Scene Completion for Safe and Efficient Exploration Mapping and Planning
41
star
96

maplab_rovio

Hard-fork of ROVIO to integrate localization.
C++
40
star
97

rtklibros

rtklib with ros interfacing and adapted feedback from external Kalman filter
C
40
star
98

libvisensor

Low level hardware driver for the visual inertial SLAM sensor.
C++
39
star
99

3dsnet

3DSNet: Unsupervised Shape-to-shape 3D Style Transfer
C++
39
star
100

mav_system_identification

Matlab scripts to perform system identification for muti-rotor systems
MATLAB
38
star