ros-planning/navigation2
Navigation2
Navigation2 is a navigation stack for ROS 2 designed for autonomous robot navigation. It provides a framework for computing optimal paths from a starting position to a goal while avoiding static and dynamic obstacles.
The system utilizes a behavior tree orchestrator to coordinate complex navigation tasks and trigger recovery actions. Its architecture is plugin-based, allowing planners, controllers, and costmap layers to be swapped at runtime without recompiling the core system.
The project covers path planning, motion control, and environmental mapping. It generates occupancy grids and costmaps to identify traversable areas and translates planned trajectories into velocity commands for physical robot tracking. Additional capabilities include trajectory smoothing and real-time collision monitoring to prevent environmental impacts.
The system manages component states through a lifecycle-managed state machine to ensure synchronized startup and shutdown.
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.
