ros-planning/navigation2
Navigation2
Navigation2 es un stack de navegación para ROS 2 diseñado para la navegación autónoma de robots. Proporciona un framework para calcular rutas óptimas desde una posición inicial hasta un objetivo, evitando obstáculos estáticos y dinámicos.
El sistema utiliza un orquestador basado en árboles de comportamiento para coordinar tareas de navegación complejas y activar acciones de recuperación. Su arquitectura es modular (basada en plugins), lo que permite intercambiar planificadores, controladores y capas de mapas de costes en tiempo de ejecución sin necesidad de recompilar el sistema principal.
El proyecto abarca la planificación de rutas, el control de movimiento y el mapeo del entorno. Genera rejillas de ocupación y mapas de costes para identificar áreas transitables y traduce las trayectorias planificadas en comandos de velocidad para el seguimiento del robot físico. Sus capacidades adicionales incluyen el suavizado de trayectorias y el monitoreo de colisiones en tiempo real para evitar impactos.
El sistema gestiona los estados de los componentes mediante una máquina de estados con ciclo de vida gestionado para asegurar un inicio y apagado sincronizados.
Features
- Autonomous Robot Navigation - Provides a complete framework for autonomous robot navigation, including path planning and obstacle avoidance.
- Robotic Behavior Execution - Executes AI-driven behavior tree logic to trigger physical robotic navigation and recovery actions.
- Behavior Trees - Implements a hierarchical behavior tree to coordinate complex navigation tasks and recovery sequences.
- Trajectory Tracking - Calculates velocity commands by comparing the current robot pose against a planned sequence of spatial coordinates.
- Occupancy Grid Generators - Generates occupancy grids that represent environmental obstacles as cost values to inform path planning.
- Ground Robot Obstacle Avoidance - Implements obstacle avoidance for ground robots using real-time sensor data and costmap generation.
- Collision Monitoring Systems - Tracks proximity to obstacles in real-time to prevent impacts and trigger safety maneuvers.
- Path Planning Algorithms - Computes optimal trajectories from start to goal while accounting for static and dynamic obstacles.
- Path Tracking - Executes path tracking and motion control to ensure the robot accurately follows planned trajectories.
- Robotics and Control - Translates planned paths into precise velocity commands and steering movements for physical robot tracking.
- Path Navigation - Computes optimal trajectories for autonomous movement between points while avoiding obstacles.
- Heuristic Search Planners - Computes optimal movement trajectories from a starting position to a goal using pathfinding algorithms.
- Robot Pathfinding - Computes optimal trajectories from a starting point to a destination while avoiding known and dynamic obstacles.
- Velocity Command Generation - Calculates the necessary velocity commands to steer the robot toward its destination along a planned path.
- Spline Trajectory Generators - Refines raw paths into kinematically feasible trajectories using spline-based interpolation to ensure fluid robot motion.
- Publish-Subscribe Messaging - Uses a publish-subscribe model to exchange sensor data and goals between decoupled planning and execution nodes.
- Robot Lifecycle Orchestration - Coordinates the initialization, active, and cleanup phases of robotic components to ensure synchronized system startup and shutdown.
- Node Runtime State Machines - Provides a standardized state machine to coordinate the deterministic startup and shutdown of software nodes.
- Plugin-Based Architectures - Employs a plugin architecture to swap planners and controllers at runtime without recompiling the core system.
- Robotics Applications - Navigation stack for ROS2.
