Regulated pure pursuit for robot path tracking

The accelerated deployment of service robots have spawned a number algorithm variations to better handle real-world conditions. Many local trajectory planning techniques been deployed on practical robot systems successfully. While most formulations Dynamic Window Approach and Model Predictive Control can progress along paths optimize for additional criteria, the use pure path tracking algorithms is still commonplace. Decades later, Pure Pursuit its variants continues be one commonly utilized classes planners. However, few proposed with schema variable linear velocities—they either assume constant velocity or fails address point at all. This paper presents variant designed heuristics regulate velocities, built atop existing Adaptive variant. Regulated makes incremental improvements state art by adjusting velocities particular focus safety in constrained partially observable spaces negotiated robots. We present experiments industrial-grade also provide high-quality reference implementation that freely included ROS 2 Nav2 framework https://github.com/ros-planning/navigation2 fast evaluation.