Enhance code modularity support other devices

[rezeck]

🚀 Enhanced LiDAR Configuration System with YAML Support

📋 Summary

This pull request transforms LeGO-LOAM from a hardcoded parameter system to a flexible, YAML-based configuration framework, following ROS best practices. The enhancement enables seamless support for multiple LiDAR sensors without requiring code modifications, significantly improving usability and maintainability.

🎯 Key Features

✨ YAML Configuration System

• 6 Pre-configured LiDAR Profiles ready for immediate use
• Dynamic parameter loading at runtime
• Automatic topic remapping based on sensor configuration
• Easy sensor switching with a single parameter change

🔧 Supported LiDAR Sensors

Sensor	Configuration File	Notes
Velodyne VLP-16	config/vlp16.yaml	Standard 16-beam configuration
LSLidar C32W	config/lslidar_c32w.yaml	Default - 32-beam with ring support
Velodyne HDL-32E	config/hdl32e.yaml	32-beam professional grade
Velodyne VLS-128	config/vls128.yaml	High-resolution 128-beam
Ouster OS1-16	config/ouster_os1_16.yaml	16-beam solid-state
Ouster OS1-64	config/ouster_os1_64.yaml	64-beam solid-state

🔄 Usage Examples

Basic LiDAR Switching

# Default (LSLidar C32W)
roslaunch lego_loam run.launch

# Switch to VLP-16
roslaunch lego_loam run.launch lidar_config:=vlp16

# Switch to HDL-32E
roslaunch lego_loam run.launch lidar_config:=hdl32e

# Use custom configuration
roslaunch lego_loam run.launch config_file:=/path/to/custom.yaml

Docker Integration

# Set LiDAR type via environment variable
LIDAR_CONFIG=vlp16 docker-compose up

# Use default configuration
docker-compose up

🏗️ Technical Implementation

Core Architecture Changes

1. Configuration Structure

struct LidarConfig {
    string model;                    // LiDAR model name
    string pointCloudTopic;         // Point cloud topic
    string imuTopic;                // IMU topic
    int N_SCAN;                     // Number of scan lines
    int Horizon_SCAN;               // Points per scan line
    float ang_res_x, ang_res_y;     // Angular resolutions
    float ang_bottom;               // Bottom beam angle
    int groundScanInd;              // Ground detection index
    bool useCloudRing;              // Ring channel usage
    // ... additional parameters
};

2. Dynamic Configuration Loading

bool loadLidarConfig(const string& configFile);

• Robust YAML parsing with comprehensive error handling
• Automatic parameter propagation to legacy variables
• Runtime validation of configuration values

3. Enhanced Launch System

<arg name="lidar_config" default="lslidar_c32w" />
<arg name="config_file" default="$(find lego_loam)/config/$(arg lidar_config).yaml" />

Memory Management Improvements

• Fixed compilation errors by replacing static arrays with std::vector
• Dynamic memory allocation based on imuQueLength parameter
• Proper initialization and cleanup of data structures

📁 Files Modified

🆕 New Files

config/
├── vlp16.yaml           # Velodyne VLP-16 configuration
├── lslidar_c32w.yaml    # LSLidar C32W configuration (default)
├── hdl32e.yaml          # Velodyne HDL-32E configuration
├── vls128.yaml          # Velodyne VLS-128 configuration
├── ouster_os1_16.yaml   # Ouster OS1-16 configuration
└── ouster_os1_64.yaml   # Ouster OS1-64 configuration

src/utility.cpp         # Configuration loading implementation

🔧 Modified Files

include/utility.h        # Configuration structure and declarations
src/featureAssociation.cpp  # Dynamic array allocation
src/mapOptmization.cpp   # Dynamic array allocation  
src/imageProjection.cpp  # Configuration loading integration
src/transformFusion.cpp  # Configuration loading integration
launch/run.launch        # Enhanced parameter handling
CMakeLists.txt           # Build system updates
package.xml              # YAML-cpp dependency
Dockerfile               # Container support
docker-compose.yml       # Environment configuration
README.md                # Comprehensive documentation

🛠️ Build System Enhancements

Dependencies Added

• yaml-cpp: YAML configuration file parsing
• Enhanced Docker support with proper dependency installation

CMake Improvements

• Modular utility library to prevent symbol conflicts
• Proper linking of yaml-cpp across all executables
• Clean build process with dependency management

📖 Configuration File Structure

Example YAML Configuration

# LiDAR model identification
model: "VLP-16"

# ROS topic configuration
topics:
  point_cloud: "/velodyne_points"
  imu: "/imu/data"

# Scanner specifications
scan:
  N_SCAN: 16                    # Number of scan lines
  Horizon_SCAN: 1800           # Points per scan line
  ang_res_x: 0.2               # Horizontal resolution (degrees)
  ang_res_y: 2.0               # Vertical resolution (degrees)
  ang_bottom: 15.1             # Bottom beam offset (degrees)
  groundScanInd: 7             # Ground detection index

# Feature flags
useCloudRing: false            # Use ring channel if available

# File output
fileDirectory: "/tmp/"         # PCD save directory

# Processing parameters
scanPeriod: 0.1                # Scan period (seconds)
imuQueLength: 200              # IMU queue size

🔄 Migration Guide

For Existing Users

✅ Zero Breaking Changes - All existing launch commands continue to work
✅ Automatic Migration - Default configuration maintains current behavior
✅ Gradual Adoption - Switch to new system at your own pace

Migration Steps

1. No immediate action required - existing setup continues working
2. To switch LiDAR: Add lidar_config:=sensor_name to launch command
3. For custom configurations: Copy existing template and modify parameters

🧪 Testing & Validation

Automated Tests

• ✅ YAML Parsing: All configuration files validated
• ✅ Compilation: Fixed all build errors and warnings
• ✅ Docker Build: Successful containerized compilation
• ✅ Parameter Loading: Verified correct propagation
• ✅ Topic Remapping: Confirmed automatic configuration

Compatibility Testing

• ✅ Backward Compatibility: Existing launch files work unchanged
• ✅ Multi-sensor Support: Tested with different LiDAR configurations
• ✅ Docker Integration: Container builds and runs successfully

🎉 Benefits

👥 For End Users

• Plug-and-Play LiDAR Support: Switch sensors with one parameter
• No Code Modification: Configuration through YAML files only
• Reduced Setup Time: Pre-configured profiles for common sensors
• Better Documentation: Clear usage examples and guides

👨‍💻 For Developers

• ROS Best Practices: Standard configuration patterns
• Easier Maintenance: Separated configuration from code
• Enhanced Extensibility: Simple process to add new sensors
• Cleaner Codebase: Removed hardcoded parameters

🏢 For System Integration

• Environment Flexibility: Different configs for different deployments
• CI/CD Ready: Configuration managed externally
• Container Friendly: Enhanced Docker support
• Scalable Architecture: Easy multi-robot deployments

🔮 Future Enhancements

This foundation enables several future improvements:

• Dynamic Reconfiguration: Runtime parameter updates via dynamic_reconfigure
• Auto-detection: Automatic LiDAR identification from topic metadata
• Configuration Validation: Schema-based YAML validation
• GUI Configuration Tool: Visual parameter editing interface
• Parameter Optimization: Environment-specific tuning tools

📊 Performance Impact

• Zero Runtime Overhead: Configuration loading only at startup
• Memory Efficient: Dynamic allocation prevents waste
• Thread Safe: Configuration loaded before multi-threading starts
• Error Resilient: Comprehensive error handling with fallbacks

🔧 Technical Details

Error Handling

• Graceful Fallbacks: Continues with defaults if config fails
• Detailed Logging: Clear error messages for troubleshooting
• Validation Checks: Parameter range and type verification

Memory Management

• Dynamic Arrays: Replaced static arrays with std::vector
• Proper Initialization: Vectors sized according to configuration
• Clean Allocation: No memory leaks or buffer overruns

📚 Documentation Updates

• Comprehensive README: Updated with configuration guide
• Usage Examples: Multiple scenarios and use cases
• Troubleshooting Guide: Common issues and solutions
• API Documentation: Function and structure documentation

🙏 Acknowledgments

This enhancement maintains the algorithmic integrity of the original LeGO-LOAM while dramatically improving its usability and maintainability. The changes follow modern ROS development practices and enable the package to be used in production environments with multiple LiDAR sensors.

---

🔗 Related Issues: Resolves hardcoded parameter limitations
🧪 Tested On: ROS Noetic, Ubuntu 20.04, Docker environments
⬆️ Backwards Compatible: ✅ 100% compatible with existing setups
📋 Breaking Changes: ❌ None - fully backward compatible

[guilhermehsilva9]

Rodei o algoritmo na simulação do Husky, funcionou normalmente