Obstacle Avoidance Algorithms

Obstacle avoidance algorithms enable autonomous systems to navigate around physical or virtual obstacles in real-time. They take sensor data (LiDAR, camera, ultrasonic) and compute steering commands that move the system toward a goal while avoiding collisions. Common algorithms include Vector Field Histogram (VFH), which builds a 2D occupancy grid and steers away from densest obstacles; Dynamic Window Approach (DWA), which predicts robot trajectories and selects collision-free ones; Rapidly-exploring Random Trees (RRT*), which sample-based path planning; and potential field methods, which treat obstacles as repulsive forces.