https://github.com/kumarrobotics/kr_param_map

A parameterized map generator for planning evaluations and benchmarking
https://github.com/kumarrobotics/kr_param_map

benchmark-framework map-generator motion-planning

Last synced: 6 months ago
JSON representation

A parameterized map generator for planning evaluations and benchmarking

• Host: GitHub
• URL: https://github.com/kumarrobotics/kr_param_map
• Owner: KumarRobotics
• License: bsd-3-clause
• Created: 2022-06-29T03:57:55.000Z (over 3 years ago)
• Default Branch: main
• Last Pushed: 2024-11-25T15:34:48.000Z (11 months ago)
• Last Synced: 2025-03-25T09:03:02.366Z (7 months ago)
• Topics: benchmark-framework, map-generator, motion-planning
• Language: C++
• Homepage:
• Size: 100 MB
• Stars: 23
• Watchers: 4
• Forks: 6
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Kumar Lab Parameterized Map

All in one! It's a revolution in map parameterization and representation for motion planning.

## About

This repo is related to our evaluation research. If this repo helps your research, please cite our paper at:

```

@INPROCEEDINGS{10610207,

  author={Shao, Yifei Simon and Wu, Yuwei and Jarin-Lipschitz, Laura and Chaudhari, Pratik and Kumar, Vijay},

  booktitle={2024 IEEE International Conference on Robotics and Automation (ICRA)}, 

  title={Design and Evaluation of Motion Planners for Quadrotors in Environments with Varying Complexities}, 

  year={2024},

  volume={},

  number={},

  pages={10033-10039},

  keywords={Measurement;Navigation;Software algorithms;Kinematics;Planning;Complexity theory;Trajectory},

  doi={10.1109/ICRA57147.2024.10610207}}

```

## 0. Setup

### 0.1 general map annotation



  



## 1. Usage

### 1.1 Grid Map Reader

It supports file format as:

- image map: .png 

- rosbag: sensor_msgs::PointCloud

- rosbag: sensor_msgs::PointCloud2

- pcd

set the map mode and file path in "read_grid_map.launch" and launch 

```

roslaunch param_env read_grid_map.launch

```

The 2D image map is converted into 3d, with z-axis having the same content. Maze images are generated with the help of [Multi Solution Maze Generator](https://github.com/shaoyifei96/multi_solution_mazegenerator)



  

  



The point cloud in bags is shown as:



  

  



You can also set the inflation (m) to inflate the grid map

```

```



  

  



The topics:

- /read_grid_map/global_cloud: publish the point clouds 

- /read_grid_map/global_gridmap: publish the center points of the grid map with inflation

### 1.2 Sctructed Map Generator

```

roslaunch param_env structure_map.launch

```

You can adjust the approximate ratio of each element (overlapping is also counting now) in the launch file 

```

```

Examples:



  

  

  

  



By increasing the occupied ratios, it's harder to generate feasible trajectories



  

  

  



You can also enable noise around the obstalce by setting:

```

  

```

The topics:

- /structure_map/global_cloud: publish the point clouds 

### 1.3 Change resolution or change map

You can also change the resolution online by publishing the resolution value to:

```

/structure_map/change_res

```

or

```

/read_grid_map/change_res

```



  

  

  

  



For read maps, you can publish all the maps by setting folder path in launch file

```

 

```

Publish true in topic

```

/structure_map/change_map

```

or

```

/read_grid_map/change_map

```

to trigger the function.

### 1.4 Dataset generation 

If you want to save map, set "dataset/save_map" to true, and set the data number in "dataset/samples_num".

```

    

    

    

    

```

It will be saved into the path you set.