SLAM Based Mobile Robot Navigation Using ROS2 — Implementation & Analysis

Abstract — SLAM Based Mobile Robot Navigation Using ROS2

Abstract. The demand for construction site automation with mobile robots is in- creasing due to its advantages in potential cost-saving, productivity, and safety. To be realistically deployed in construction sites, mobile robots must be capable of navigating in unstructured and cluttered environments. Furthermore, mobile robots should recog- nize both static and dynamic obstacles to determine drivable paths. However, existing robot navigation methods are not suitable for construction applications due to the chal- lenging environmental conditions in construction sites. This study introduces an auton- omous as-is 3D spatial data collection and perception method for mobile robots specifically aimed for construction job sites with many spatial uncertainties. The pro- posed Simultaneous Localization and Mapping (SLAM)-based navigation and object recognition methods were implemented and tested with a custom-designed mobile ro- bot platform, Ground Robot of Mapping Infrastructure (GRoMI), which uses multiple laser scanners and a camera to sense and build a 3D environment map. Since SLAM, it self, did not detect uneven surface conditions and spatiotemporal objects on the ground, an obstacle detection algorithm was developed to recognize and avoid obstacles and the highly uneven terrain in real time. Given the 3D real-time scan map generated by 3D laser scanners, a path-finding algorithm was developed for autonomous navigation in an unknown environment with obstacles. Overall, the 3D color-mapped point clouds of construction sites generated by GRoMI were of sufficient quality to be used for many construction management applications such as construction progress monitoring, safety hazard identification, and defect detection.

Operational Methodology & Framework

Keywords: Mobile robot, Navigation, SLAM, Object recognition, Unstructured environment, Construction, Point clouds Introduction The current commercial 3D laser scanning solutions for the construction industries can collect millions of three-dimensional points in a short period of time accurately and safely under stationary conditions. However, the post-processing, which is point cloud registration process to combine each point cloud from different scan locations into one coordinate system, is still a labor-intensive and time-consuming process. First, multiple scans from different scan locations are manually gathered by the operator with targets which are marked correspondences in the scan area to find common points between point clouds. Second, it can cause errors in placing and relocating targets. Third, these collected data are registered with matching common features or targets manually. How- ever, the operator does not know until the end of the registration process that the col- lected data contain imperfections due to incomplete scans, hindering structures, absence of targets and lack of common features for registration in many cases. This is because current laser scanning methods provide limited feedback to the operator during the scan process and registration process . As like this, the current quality control methods at a job site rely heavily on manual inspection which is labor-intensive, tedious, and error- prone. Also, the raw data collected from construction sites are often unstructured and poorly organized which results in time and cost overheads in interpreting the data as well as delays in sharing the relevant information to stakeholders . Therefore, robot technology has the potential to reduce construction and maintenance costs and improve productivity, quality, and safety. The automated data collection of construction site conditions by a mobile robot would provide an attractive alternative for the execution of routine work tasks at a job site [3, 4].

Experimental Findings & Validation

To be a practical solution for construction applications, an autonomous mobile robot must have the capability to safely move around an unstructured and cluttered environ- ment with many spatiotemporal objects and obstacles which continuously change the environment conditions (i.e., uneven ground surface, equipment, materials, soil stock- piles, temporary structures). Robot’s obstacle recognition and avoidance mechanisms should be fast, robust and not solely dependent on the as-designed construction infor- mation of the environment because quickly changing as-is conditions of a construction job site are different from as-designed conditions in most cases. Several methods exist in the literature for robot navigation in other domains which work by observing an ex- isting map and planning the robot’s every move in advance . However, the existing methods are not suitable for construction applications due to the aforementioned construction site’s unique characteristics.

Scope & Practical Applications

• Real-time automated control and signal telemetry
• Embedded verification testbeds for institutional research
• High-throughput data streaming and anomaly alerting
• Performance optimization and resource consumption minimization
• Full integration ready for IEEE dissertation submission