Obstacle Detection

📚 Contents

• Description
• Architecture
• Interfaces
• Installation
• Usage
• License

🧠 Description

The Obstacle Detection node receives its input from the 2D YDLidar sensor through the /scan topic. This topic publishes messages of type sensor_msgs/LaserScan.msg, which contain a series of distance measurements (ranges) along with corresponding angular information. These laser scan messages provide data by measuring the distance between the lidar and nearby objects at various angles. This rich set of data serves as the foundation for the obstacle detection process.

The primary functionality of the Obstacle Detection node is to interpret the Lidar data to detect and identify obstacles in the vehicle's surroundings. It processes the incoming LaserScan messages to analyze range values and detect significant discontinuities or clusters that represent physical obstacles. The component calculates the distance of these obstacles relative to the ego vehicle (base_link), filters out noise or irrelevant data, and organizes the detected objects into a structured format. This process gives the high-level information about the presence, left, right edge coordinates, and distance of obstacles around the ego vehicle.

As a result of its processing, the Obstacle Detection node publishes information on the /obstacles topic using messages of type ObstacleDetectionArray.msg. These output messages contain an array of detected obstacles along with details such as their left, right edge coordinates and distances from the ego vehicle.

Functionality:

Extract all the obstacles from the 2D LiDAR measurements

Activity Diagram	/scan and /filtered_scan

This activity diagram illustrates a 2D LiDAR-based obstacle detection pipeline. The process begins by subscribing to the /scan topic to receive raw LiDAR data, which is then filtered to retain only ranges between 0.5 m and 2.5 m. The filtered scan is published in topic /filtered for visualization. The corresponding Cartesian coordinates (x, y) are computed from the filtered range and angle measurements. Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering is applied to group nearby points into distinct clusters. For each cluster, the coordinates are transformed from the laser_frame to the base_link frame. Within the transformed cluster, the left and right edges are determined based on the maximum and minimum y values, and the centroid is calculated to estimate the distance to the obstacle from base_link. The obstacle’s ID, left & right edge coordinates, and distance are stored, and once all clusters are processed, the compiled obstacle data array is published on the /obstacles topic.

🧩 Architecture

graph LR

    subgraph Input topics
        LIDAR["/scan"]:::grayEllipse
    end
    
    OD["obstacle_detection"]:::cyanEllipse
    
    subgraph Output topics
        EM["/obstacles"]:::grayEllipse
        FS["/filtered_scan"]:::grayEllipse
    end

    LIDAR --> OD
    OD --> EM
    OD --> FS

    %% Ellipse shape class
    classDef soft_rectangle stroke:#FFFFFF,rx:20,ry:20;
    classDef component font-weight:bold,stroke-width:2px;

    %% Cyan for path planner
    classDef cyanEllipse fill:#00CED1, color:#000000;

    %% Gray for others
    classDef grayEllipse fill:#D3D3D3, color:#000000;

    %% Apply ellipse shape to each node
    class OD component;
    class EM soft_rectangle;
    class FS soft_rectangle;
    class LIDAR soft_rectangle;

Loading

🔌 Interfaces

Topics:

Name	IO	Type	Description
/scan	Input	sensor_msgs/msg/LaserScan.msg	Provides angles and ranges
/obstacles	Output	custom_msgs/msg/ObstacleDetectionArray.msg	Provides obstacles ID, left & right edge coordinates and distance from base_link frame
/filtered_scan	Output	sensor_msgs/msg/LaserScan.msg	Provides filtered ranges and angles for visualization

Custom messages:

Message: ObstacleDetectionArray.msg

Name	Type	Description
obstacles	ObstacleDetectionData[]	Array of obstacles with unique ID, position and distance

Message: ObstacleDetectionData.msg

Name	Type	Description
header	std_msgs/Header	for time and frame id
obstacle_id	uint8	unique ID for obstacle
distance	float32	distance between obstacle and base_link
l_x	float32	left edge x coordinate of obstacle
l_y	float32	left edge y coordinate of obstacle
r_x	float32	right edge x coordinate of obstacle
r_y	float32	right edge y coordinate of obstacle

🛠️ Installation

1. Create workspace, src and go to src

mkdir temp_ws
cd temp_ws
mkdir src
cd src

2. Clone component repository

git clone https://github.com/surendrakoganti/obstacle_detection.git

3. Clone custom messages repository

git clone https://github.com/surendrakoganti/custom_msgs.git

4. Return to workspace and build the packages

cd ..
colcon build

5. Source the setup files

source install/setup.bash

▶️ Usage

Run the node:

ros2 launch obstacle_detection obstacle_detection_launch.py

License

Licensed under the Apache 2.0 License. See LICENSE for details.