Shape-Aware and G 2 Continuous Path Planning Based on Bidirectional Hybrid A ∗ for Car-Like Vehicles

This paper presents modified hybrid A ∗ algorithms to facilitate more efficient path finding around narrow passages under the shape and kinematic constraints of a car-like vehicle. First, we propose spline- and bidirectional-search-based hybrid A ∗ using G 2 continuous motion primitives with multiple turning radii depending on cubic Bezier curves. In addition, we present a heuristic applying the vector field histogram to find a path around narrow passages with efficiency rather than optimization. We demonstrate the benefits of our method through simulations and experimental results using an autonomous ground vehicle in environments with narrow passages.