top of page

KEY PUBLICATIONS

This page contains a list of my academic and non academic publications. For a full list of my academic publications you can check out my google scholar.

June 10, 2023

BIM-GNN: SEMANTIC AWARE QUALITY ASSESSMENTS OF BUILDING CONSTRUCTION

Automated construction quality assessments often face challenges due to noise and occlusions in on-site data when independently comparing elements in as-designed building information models (ad-BIMs) with registered 3D point clouds. We propose BIM-Graph Neural Network (BIM-GNN), a learning-based approach that leverages semantics in ad-BIM to enhance element-wise quality assessments. BIM-GNN transforms ad-BIM into a graph, where nodes represent BIM objects and edges capture their topological and spatial relationships. The node features incorporate ad-BIM semantics, scan alignments, and graph structure information, while edge attributes describe relationship types. BIM-GNN employs transductive learning to classify nodes as verified, deviated, missing, or no-data, capturing project-level context. Experimental results show that BIM-GNN outperforms baseline machine learning models.

Sept 11, 2021

SYNCHROSEM: MULTI-MODAL SEMANTIC SLAM FOR LVI SYSTEMS

SynchroSem is a loosely coupled multi-modal semantic SLAM approach for synchronized Lidar-Visual-Inertial (LVI) systems. Our system integrates pose measurements from two different tightly coupled odometry systems, Visual-Inertial (VI) and Lidar-Inertial (LI). These are loosely coupled by means of a pose graph representation that enables a robust late-fusion of pose measurements, being resilient to geometric and visual degeneracy. The multi-modal semantic feature extraction allows the global optimization system to reduce long term errors via loop closure, achieving state-of-the-art performance.

April 14, 2021

ADVANCING CONSTRUCTION: DIGITAL TOOLS FOR SITE EFFICIENCY

The combination of outdated tools and processes prevents construction site managers and other stakeholders from efficiently tracking progress and measuring quality on their construction projects. This leads to delays and budget overruns caused by rework. Digital tools using AI and ML can measure, track, and analyze construction processes to enable efficient process control, thereby reducing waste and inefficiency.

August 11, 2017

INTERACTIVE PERCEPTION

Recent approaches in robot perception follow the insight that perception is facilitated by interaction with the environment. These approaches are subsumed under the term Interactive Perception (IP). This view of perception provides the following benefits. First, interaction with the environment creates a rich sensory signal that would otherwise not be present. Second, knowledge of the regularity in the combined space of sensory data and action parameters facilitates the prediction and interpretation of the sensory signal.

bottom of page