Date: 19th August, 2026
Linkedln Post: https://lnkd.in/p/ennAcrgT
Written by: Anyanwuocha-Collins Chibuike Innocent
<aside>
</aside>
A.L.T.A.R. (Autonomous Locator, Tracker and Analysis Rover) is an autonomous robotic inspection system developed to improve the detection and assessment of defects in concrete structures. The project addresses a persistent challenge in the construction industry, where hidden defects can contribute to poor-quality construction, costly repairs, project delays, and, in severe cases, structural failure and building collapse.
Conventional methods such as hammer sounding and core sampling can be subjective, destructive, time-consuming, and limited in coverage. A.L.T.A.R. provides a locally engineered, non-destructive alternative by combining Impact-Echo Analysis, Infrared Thermal Imaging, and AI-based Computer Vision to detect internal and surface defects such as voids, honeycombing, and cracks.
The system integrates mechanical, electrical, sensing, computing, and software subsystems into a single autonomous platform. Its complete 3D CAD assembly was developed in SolidWorks, with custom-designed components and mounting structures created specifically for the system to ensure proper fit, clearance, and integration.
The project was also presented at the Julius Berger Innovation Challenge (JBIC), where A.L.T.A.R. emerged as the first-place winner, providing external recognition of its innovation and potential for construction applications.
The resulting prototype demonstrates how robotics, CAD engineering, non-destructive testing, and artificial intelligence can be combined to improve structural inspection and support safer, more efficient construction.
<aside>
</aside>
Concrete structures are expected to meet specific quality and performance requirements throughout their service life. However, defects can be introduced during material preparation, transportation, placement, compaction, curing, or subsequent construction activities. Some of these defects remain hidden beneath the concrete surface and may not become obvious until they have already affected structural performance.
Defects such as honeycombing, internal voids, cracks, and areas of poor consolidation can compromise the quality of a concrete element. When these conditions are not identified early, corrective action becomes more difficult and expensive.
This creates a significant inspection challenge. Once concrete has been cast, engineers cannot simply observe its entire internal condition visually. Conventional approaches therefore rely on indirect assessment or localized testing.
Hammer sounding, for example, depends heavily on the experience and judgement of the person carrying out the inspection. Core sampling provides physical evidence of the concrete condition, but it is destructive and only examines localized regions. As a result, inspecting a large structure through conventional techniques can require significant time, labour, and cost while still leaving substantial areas unexamined.
A.L.T.A.R. was developed to address this gap by introducing automated, non-destructive and data-driven inspection into the construction workflow.
<aside>
</aside>
A.L.T.A.R. is designed as more than a mobile sensor platform. It is a complete inspection system in which the robot's mobility, sensing hardware, processing systems, and software work together to transform physical inspection data into engineering information.
The basic operating concept follows a sequence:
Position → Scan → Sense → Process → Analyse → Assess → Report
The robot is positioned on the concrete surface requiring inspection and moves across the designated area while collecting information from its integrated inspection systems. The collected data is processed using onboard computing resources, including Edge AI capabilities, before being interpreted to determine the condition of the inspected region.