Giters

AUST-Hansen / Ground-Challenge

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Ground-Challenge: A Multi-sensor SLAM Dataset Focusing on Corner Cases for Ground Robots

Figure 1. Different corner cases for SLAM

Notice:

1. All the rosbag files have been released now, while the GT, calibration files and robot size parameters will be made public upon paper acceptance. All the experiment results in the paper can be reproduced.

2. To get the calibration and GT files in advance, you can contact us at 1195391308@qq.com. If our work is helpful for your research, please follow us and give a star

ABSTRACT:

We introduce Ground-Challenge: a novel dataset collected by a ground robot with multiple sensors including an RGB-D camera, an inertial measurement unit (IMU), a wheel odometer and a 3D LiDAR to support the research on corner cases of visual SLAM systems. Our dataset comprises 36 trajectories with diverse corner cases such as aggressive motion, severe occlusion, changing illumination, few textures, pure rotation, motion blur, wheel suspension, etc. Some state-of-the-art SLAM algorithms are tested on our dataset, showing that these systems are seriously drifting and even failing on specific sequences. We will release the dataset and relevant materials upon paper publication to benefit the research community.

MAIN CONTRIBUTIONS:

  • We collect a novel visual SLAM dataset for ground robots with a rich pool of sensors in diverse environments both indoors and outdoors. Particularly, the dataset covers a series of challenging sequences for sensor failures and specific movement patterns.
  • State-of-the-art SLAM algorithms of different settings are tested on our benchmark. And the results indicate these systems are not robust enough for situations such as sensor failures.
  • To facilitate the research on corner cases of robot navigation, we will release the dataset with ground truth trajectories and the configuration file of each tested algorithm upon paper publication.

1.SENSOR SETUP

1.1 Acquisition Platform

The ground robot is given below. The unit of the figures is centimeter.

Figure 2. The data capture robot.

1.2 Sensor parameters

All the sensors and track devices and their most important parameters are listed as below:

  • LIDAR Velodyne VLP-16, 360 Horizontal Field of View (FOV),-30 to +10 vertical FOV,10Hz,Max Range 200 m,Range Resolution 3 cm, Horizontal Angular Resolution 0.2°.

  • V-I Sensor,Realsense d435i,RGB/Depth 640*480,69H-FOV,42.5V-FOV,15Hz;IMU 6-axix, 200Hz

  • IMU,Xsens Mti-300,9-axis,400Hz;

  • Wheel Odometer,AgileX,2D,25Hz;

The rostopics of our rosbag sequences are listed as follows:

  • LIDAR: /velodyne_points

  • V-I Sensor:
    /camera/color/image_raw ,
    /camera/depth/image_raw ,
    /camera/imu

  • IMU: /imu/data

  • Wheel Odometer: /odom

2.DATASET SEQUENCES

An overview of Ground-Challenge is given in the table below:

Scenario Darkroom Occlusion Office Room Wall Motionblur Hall Loop Roughroad Corridor Rotation Static Slope TOTAL
Number 3 4 3 3 3 3 3 2 3 2 3 2 2 36
Dist/m 92.0 273.8 75.5 102.1 86.7 166.6 236.3 371.8 68.1 164.3 12.4 1.9 128.5 1780.0
Duration/s 203.6 334.2 164.0 154.7 189.3 145.5 302.4 332.7 186.3 198.1 183.2 92.6 195.0 2681.6
Size/GB 6.1 9.9 4.7 4.6 5.6 4.3 8.7 9.9 5.4 5.8 5.4 2.7 5.7 78.8

2.1 Visual Challenges

Sequence Name Total Size Duration Features Rosbag GT
Darkroom1 2.9g 100s slight light, going into a room Rosbag GT
Darkroom2 2.3g 76s sharp turn Rosbag GT
Darkroom3 1.9g 64s slight light Rosbag GT
Occlusion1 2.9g 97s moving feet, far features Rosbag GT
Occlusion2 3.2g 108s hand occlusion Rosbag GT
Occlusion3 2.6g 89s hand occlusion Rosbag GT
Occlusion4 1.2g 40s complete occlusion Rosbag GT
Office1 1.3g 46s exposure change Rosbag GT
Office2 1.9g 66s going into a dark room Rosbag GT
Office3 1.5g 52s office Rosbag GT
Room1 1.3g 46s exposure change Rosbag GT
Room2 1.9g 66s going into a dark room Rosbag GT
Room3 1.5g 52s office Rosbag GT
Motionblur1 1.5g 52s aggressive motion Rosbag GT
Motionblur2 1.6g 54s aggressive motion Rosbag GT
Motionblur3 1.2g 40s aggressive motion Rosbag GT
Wall1 1.7g 59s wall in a corridor Rosbag GT
Wall2 2.0g 66s wall in a big hall Rosbag GT
Wall3 3.9g 65s wall in a corridor Rosbag GT

2.2 Wheel Challenge

Sequence Name Total Size Duration Features Rosbag GT
Hall1 2.6g 91s slippery ground, a reflective surface Rosbag GT
Hall2 3.2g 110s slippery ground, a reflective surface Rosbag GT
Hall3 2.9g 101s slippery ground, walking human Rosbag GT
Loop1 4.1g 97s moving feet, far features Rosbag GT
Loop2 5.8g 137s hand occlusion Rosbag GT
Roughroad1 2.2g 75s rough road Rosbag GT
Roughroad2 1.5g 52s rough road Rosbag GT
Roughroad3 1.8g 59s rough road Rosbag GT

2.3 Specific Movement Patterns

Sequence Name Total Size Duration Features Rosbag GT
Corridor1 2.9g 100s zigzag, long corridor Rosbag GT
Corridor2 2.9g 98s straight forward, long corridor Rosbag GT
Rotation1 1.6g 53s moving feet, far features Rosbag GT
Rotation2 2.1g 73s hand occlusion Rosbag GT
Rotation3 1.7g 57s rough road Rosbag GT
Static1 1.6g 56s rough road Rosbag GT
Static2 1.1g 37s rough road Rosbag GT
Slope1 2.8g 96s slope Rosbag GT
Slope2 2.9g 99s slope Rosbag GT

3. CONFIGURERATION FILES

For convenience of evaluation, we provide configuration files of some well-known SLAM systems as below:

To be uploaded

4. CALIBRATION FILES

To be uploaded

About