About me

I’m Sali Perja, a BEng Mechatronics and Robotics student at MDX University with a precision fabrication and design engineering background. My professional experience as a staircase designer and fabricator deepened my understanding of structural design, CAD modelling, and manufacturing processes. Over the years, I have worked on custom-built staircases, overseeing the entire process from design to fabrication and installation. Beyond my professional work, I am passionate about robotics, automation, and AI-driven solutions that can improve industrial efficiency. My expertise spans 3D CAD software (SolidWorks, AutoCAD, Fusion 360), CNC machining, and Lidar-based data processing. This blog will document my journey in developing an innovative Lidar-based AI System for Construction Site Verification and Floor Plan Comparison.

Motivation and Main Goal

The construction industry relies heavily on manual inspections to ensure building structures align with their original floor plans. However, these traditional verification methods are time-consuming, labour-intensive, and prone to human error, leading to inefficiencies and increased costs.

My project aims to develop a Lidar-based system capable of autonomously scanning construction sites and comparing real-world structures with DXF floor plans. This system will:

• Capture and process 3D point cloud data using a Velodyne Puck LITE Lidar sensor.

• Filter and segment structural elements (walls, pillars, permanent features) while removing irrelevant objects.

• Compare the scanned data against floor plans using feature detection and edge recognition.

• For further analysis, export results in various formats, including DXF, PCD, and 2D plan views.

By automating this process, I aim to create a faster, more accurate, cost-effective verification method that can be integrated into modern construction workflows. While my focus will primarily be developing the software logic, this project could pave the way for future real-world applications, including robotic deployment for automated scanning.

What to Expect in This Blog

What to Expect in This Blog This blog will serve as a live record of my project's progress, challenges, and key learnings. You can expect to find:

• Project Progress: Weekly updates on my development milestones.

• Technical Insights: How I integrate Lidar and automation for construction verification.

• Challenges & Solutions: The obstacles I face and the solutions I develop.

• Simulation and Testing: Results from real-world DXF floor plan comparisons.

• Final Outcome: Demonstrate the system’s functionality, including export capabilities and accuracy assessment.

I hope this blog is an insightful resource for those interested in robotics, applications in construction, and automation technologies. Let’s innovate and optimise construction verification together!