Path Planning of Two-Wheeled Self-Balancing Robots in Static Obstacle Scenarios Based on Improved A* Algorithm

Two-wheeled self-balancing robots have great application potential in indoor inspection and home services due to their compact size and high flexibility. However, their mechanical and kinematic characteristics—narrow wheel track and high center of gravity causing tipping instability—impose stringent requirements on path safety and smoothness that the traditional A* algorithm cannot meet. To address this, this study proposes an improved A* algorithm integrating obstacle expansion and moving average smoothing, solving collision risks from body size and motion instability from unsmooth paths. This study built a Python-based static obstacle simulation platform, with comparative verification in simple and complex scenarios. Results show the improved algorithm ensures a minimum 10 cm safe distance from obstacles, generates smoother paths with fewer inflection points and continuous curvature changes, and avoids collisions. This study provides a safe, smooth path planning solution for two-wheeled self-balancing robots in indoor static obstacle environments.