Start Date
1-5-2026 12:00 PM
End Date
1-5-2026 1:00 PM
Description
This project presents the design, implementation, and evaluation of an autonomous indoor mobile robot capable of simultaneous localization and mapping (SLAM) using a ROS 2-based architecture on resource-constrained embedded hardware. The system is built on a Yahboom ROSMASTER X3 platform equipped with a Slamtec RPLidar A1, IMU, and wheel odometry, running ROS 2 Foxy on an NVIDIA Jetson Nano 4GB inside a Docker container. The primary objective is to enable the robot to autonomously map an indoor hallway environment while navigating without human intervention.
The mapping pipeline integrates slam_toolbox for occupancy grid generation, with loop closure enabled and parameters tuned for the platform's motion constraints and LiDAR update rate. A LiDAR-based reactive wall-following controller was developed and validated as the primary autonomous navigation behavior, using a proportional-derivative (PD) state machine with eight operating modes: wall follow, corner wrap, turn, settle, search, explore, recovery backup, and recovery turn. A key hardware challenge was resolved by identifying a 180-degree mounting offset on the RPLidar A1, which required a corrective angle transformation applied via a static TF broadcast in the robot's coordinate frame. An additional hardware complication arose from a screen assembly factory-mounted on the robot, which introduced LiDAR occlusion artifacts and map distortion at corners. Removing the screen resolved the occlusion issue but shifted the robot's weight distribution, causing lateral drift during turns. An automated startup calibration routine was developed to dynamically set the target wall-following distance from live sensor data rather than relying on hardcoded parameters. A frontier exploration node was also implemented to extend autonomous coverage beyond wall-following range, identifying free-space boundaries in the occupancy grid, clustering and ranking frontiers by distance, and publishing navigation goals to the wall follower via a dedicated topic.
Occupancy grid maps of an indoor hallway were successfully generated across multiple runs at 0.05 m/cell resolution, with the final map clearly resolving corridor geometry including a T-junction. The full autonomy stack — bringup, SLAM, wall follower, and frontier explorer — is integrated into a single launch file. All robot-dependent work was completed prior to the April 30, 2026 hardware return deadline. Post-deadline work includes quantitative map quality analysis against a manually measured ground-truth floor plan, localization strategy comparison, and computational load benchmarking across configurations.
This work demonstrates that a low-cost, Jetson-powered mobile robot can achieve stable autonomous navigation and real-time SLAM within the constraints of limited onboard compute, and contributes a replicable open-source pipeline for indoor occupancy grid mapping on the ROS 2 Foxy ecosystem.
Autonomous Indoor Mapping and Navigation Using ROS2-Based SLAM on a Jetson-Powered Mobile Robot
This project presents the design, implementation, and evaluation of an autonomous indoor mobile robot capable of simultaneous localization and mapping (SLAM) using a ROS 2-based architecture on resource-constrained embedded hardware. The system is built on a Yahboom ROSMASTER X3 platform equipped with a Slamtec RPLidar A1, IMU, and wheel odometry, running ROS 2 Foxy on an NVIDIA Jetson Nano 4GB inside a Docker container. The primary objective is to enable the robot to autonomously map an indoor hallway environment while navigating without human intervention.
The mapping pipeline integrates slam_toolbox for occupancy grid generation, with loop closure enabled and parameters tuned for the platform's motion constraints and LiDAR update rate. A LiDAR-based reactive wall-following controller was developed and validated as the primary autonomous navigation behavior, using a proportional-derivative (PD) state machine with eight operating modes: wall follow, corner wrap, turn, settle, search, explore, recovery backup, and recovery turn. A key hardware challenge was resolved by identifying a 180-degree mounting offset on the RPLidar A1, which required a corrective angle transformation applied via a static TF broadcast in the robot's coordinate frame. An additional hardware complication arose from a screen assembly factory-mounted on the robot, which introduced LiDAR occlusion artifacts and map distortion at corners. Removing the screen resolved the occlusion issue but shifted the robot's weight distribution, causing lateral drift during turns. An automated startup calibration routine was developed to dynamically set the target wall-following distance from live sensor data rather than relying on hardcoded parameters. A frontier exploration node was also implemented to extend autonomous coverage beyond wall-following range, identifying free-space boundaries in the occupancy grid, clustering and ranking frontiers by distance, and publishing navigation goals to the wall follower via a dedicated topic.
Occupancy grid maps of an indoor hallway were successfully generated across multiple runs at 0.05 m/cell resolution, with the final map clearly resolving corridor geometry including a T-junction. The full autonomy stack — bringup, SLAM, wall follower, and frontier explorer — is integrated into a single launch file. All robot-dependent work was completed prior to the April 30, 2026 hardware return deadline. Post-deadline work includes quantitative map quality analysis against a manually measured ground-truth floor plan, localization strategy comparison, and computational load benchmarking across configurations.
This work demonstrates that a low-cost, Jetson-powered mobile robot can achieve stable autonomous navigation and real-time SLAM within the constraints of limited onboard compute, and contributes a replicable open-source pipeline for indoor occupancy grid mapping on the ROS 2 Foxy ecosystem.