Regulated Pure Pursuit for Robot Path Tracking

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

• The paper introduces Regulated Pure Pursuit which dynamically adjusts robot speeds based on path curvature and obstacle proximity to enhance safety and accuracy.
• It employs curvature and proximity heuristics that significantly reduce path tracking errors, achieving a mean error of 0.03 m compared to 0.19 m for standard methods.
• The algorithm's integration in ROS 2 Nav2 and successful experiments in real-world, constrained environments highlight its practical impact.

Regulated Pure Pursuit for Robot Path Tracking

The paper "Regulated Pure Pursuit for Robot Path Tracking" presents an enhanced version of the Pure Pursuit algorithm, focusing on improving the safety and reliability of path tracking for service and industrial robots. This variant, known as Regulated Pure Pursuit (RPP), introduces mechanisms to adjust linear velocities dynamically, which is critical in constrained environments that are only partially observable.

Context and Background

Pure Pursuit is a well-established path tracking algorithm widely applied due to its simplicity and effectiveness. Historically, these algorithms, including derivatives like Dynamic Window Approach (DWA) and Model Predictive Control (MPC), have emphasized reliability and versatility across various robotic applications. However, a limitation persists in conventional Pure Pursuit formulations which often assume constant velocities, inadequate for real-world deployment where safety considerations demand variable speeds in dynamic environments.

Contribution of the Paper

Regulated Pure Pursuit represents an incremental advancement over existing Adaptive Pure Pursuit (APP) algorithms by incorporating velocity regulation heuristics. This includes slowing the robot during sharp turns and when approaching obstacles. These heuristic adjustments are mathematically tuned to enhance both safety and operability.

1. Curvature Heuristic: Adjusts speed based on path curvature, which is particularly useful for navigating tight spaces or when the path's visibility is limited.
2. Proximity Heuristic: Slows the robot as it comes close to obstacles, improving safety in environments populated with dynamic agents like humans.

Additionally, RPP introduces preemptive collision detection, enhancing the robot's ability to navigate safely in real-time.

Experimental Validation

The authors conducted several experiments to validate the efficacy of RPP:

• Path Tracking Experiment: This simulation demonstrated RPP's ability to minimize path tracking errors, showing a mean error of 0.03 meters compared to 0.19 meters for standard Pure Pursuit.
• Blind Turning and Confined Corridor Experiments: RPP improved the robot's ability to handle blind corners and navigate narrow corridors safely, reducing the risk of collisions through dynamic speed adjustments.
• Full-System Experiment: Conducted in a complex indoor environment, results showed RPP's safety improvements without sacrificing overall system performance.

Implications

The research advances path tracking methods for mobile robots, crucial for real-world applications in industries ranging from warehousing to retail. By enhancing both path precision and safety, RPP is particularly well-suited for environments where robots operate in close proximity to humans or other obstacles. The availability of a high-quality implementation in the ROS 2 Nav2 framework significantly lowers the barrier for adoption, allowing for immediate practical application by the robotics community.

Future Directions

The development of RPP paves the way for further exploration into integrating dynamic constraints more deeply, enhancing kinematic feasibility, especially for Ackermann-steering vehicles. As robotics technology continues to evolve, ensuring adaptability in complex, real-world environments remains a critical area for ongoing research and development.

In conclusion, Regulated Pure Pursuit provides a robust, practical improvement over traditional path tracking algorithms, offering a balanced approach to safety and performance that aligns well with the growing demands of contemporary robotic applications.