SMAC Planner Hybrid-A*

Source: ros2-copilot-skills Hybrid-A* skill

Why This Matters

Hybrid-A* matters when the robot's orientation and turning constraints materially shape whether a path is actually useful. For ackermann-like or constrained-motion vehicles, this can make the difference between a path that looks legal on a grid and a path the robot can actually follow well.

Distilled Takeaways

• Hybrid-A* planning explicitly incorporates heading and motion constraints into global planning.
• It is most justified when simpler 2D planners produce paths that are hard for the vehicle to execute.
• The added complexity only pays off when the vehicle model and environment demand it.
• Planner realism and controller compatibility should be judged together.

Practical Guidance

• Choose Hybrid-A* when the vehicle is not well served by unconstrained 2D path planning.
• Compare path-following outcomes, not just planner output visuals.
• Keep vehicle limits and footprint assumptions realistic.
• Test in maps with the tight turns and directional constraints that motivated the choice.

Corroborating References

• Nav2 planner plugins docs
• Nav2 tuning guide

When to Read the Original Source

Go to the original skill when you want the Hybrid-A* specific configuration and the practical tradeoffs against lattice and simpler 2D planning.