Top AI Repos β open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Real-time LIDAR-based Urban Road and Sidewalk detection for Autonomous Vehicles π
| Date | Stars |
|---|---|
| 2026-07-24 | 356 |
| 2026-07-25 | 356 |
| 2026-07-28 | 356 |
| 2026-07-30 | 356 |
| 2026-08-06 | 356 |
Today
β stars today
This week
β stars this week
This month
β stars this month
Momentum
0.0
growth rate 0.00%/day
# `urban_road_filter`: a real-time LIDAR-based urban road and sidewalk detection algorithm for autonomous vehicles
<img src="img/urban_road_filter_anim01.gif" height=620/> <img src="img/urban_road_filter_static01.png" height=620/>
# Project site: [jkk-research.github.io/urban_road_filter](https://jkk-research.github.io/urban_road_filter/)
# Dependency
- [ROS 2](https://docs.ros.org/en/humble/index.html) (tested with [](https://docs.ros.org/en/humble/) and [](https://docs.ros.org/en/jazzy/))
- [PCL](https://pointclouds.org/)
# Install (download and build)
Use the following commands to download and compile the package.
```bash
cd ~/ros2_ws/src
```
```bash
git clone https://github.com/jkk-research/urban_road_filter -b ros2
```
And build:
```bash
cd ~/ros2_ws
```
```bash
colcon build --packages-select urban_road_filter --symlink-install --cmake-args -DCMAKE_BUILD_TYPE=Release
```
# Getting started
Issue the following commands to start ROS 2, play sample data, and start the algorithm with visualization.
In a **new terminal** start ROS 2:
Don't forget to source your workspace first
<details>
source ~/ros2_ws/install/setup.bash
</details>
```bash
ros2 launch urban_road_filter urban_road_filter.launch.py
```
# Features
The `urban_road_filter` package provides the following features:
1. **Detection Methods**:
- `x_zero_method`: Detects roadside points by analyzing the angle between three points while keeping the X-coordinate constant.
- `z_zero_method`: Detects roadside points by analyzing the angle between two vectors while keeping the Z-coordinate constant.
- `star_shaped_method`: Uses a star-shaped algorithm to detect roadside points within a sector.
2. **Configurable Parameters**:
- Detection area dimensions (`min_x`, `max_x`, `min_y`, `max_y`, `min_z`, `max_z`).
- LIDAR vertical resolution (`interval`).
- Curb detection parameters (`curb_height`, `curb_points`).
- Polygon simplification options (`polysimp_allow`, `polysimp`, `polyz`).
3. **ROS Topics**:
- `/road`: Filtered points representing the drivable road.
- `/curb`: Filtered points representing curbs.
- `/roi`: Filtered points within the region of interest.
- `/road_marker`: Visualization markers for detected roads.
4. **Performance Metrics**:
- Publishes statistics about the segmentation results, including the number of points classified as road, curb, or non-road.
# Configuration
The behavior of the `urban_road_filter` node can be customized using the `parameters.yaml` file. Key parameters include:
- `fixed_frame`: The fixed frame for the LIDAR data.
- `topic_name`: The topic to subscribe to for point cloud data.
- `x_zero_method`, `z_zero_method`, `star_shaped_method`: Enable or disable specific detection methods.
- `interval`: Acceptable interval for the LIDAR's vertical angular resolution.
- `curb_height`, `curb_points`: Parameters for curb detection.
- `polysimp_allow`, `polysimp`, `polyz`: Parameters for polygon simplification.
# Cite & paper
If you use any of this code, please consider citing the [paper](https://www.mdpi.com/1424-8220/22/1/194):
```bibtex
@Article{roadfilt2022horv,
title = {Real-Time LIDAR-Based Urban Road and Sidewalk Detection for Autonomous Vehicles},
author = {HorvΓ‘th, ErnΕ and Pozna, Claudiu and Unger, MiklΓ³s},
journal = {Sensors},
volume = {22},
year = {2022},
number = {1},
url = {https://www.mdpi.com/1424-8220/22/1/194},
issn = {1424-8220},
doi = {10.3390/s22010194}
}
```
# Related solutions
- [`points_preprocessor`](https://github.com/Autoware-AI/core_perception/tree/master/points_preprocessor) `ray_ground_filter` and `ring_ground_filter` (ROS)
- [`linefit_ground_segmentation`](https://github.com/lorenwel/linefit_ground_segmentation) (ROS)
- [`curb_detection`](https://github.com/linyliny/curb_detection) (ROS)
- [`Excerpt of 5,545 characters
Read on GitHubWould you bet a product on this? Bounded 0β100 and slow moving.
matched fp:be1703947beb2954, topic:ros, topic:ros2, topic:autonomous-driving