Investigation into a pick and place operation for onsite construction application

Bricklaying is one of the most accident-prone professions in the construction industry, with heavy and non-ergonomic lifting being a leading cause of sick leave and back problems. Increased automation is expected to reduce the number of work-related accidents while simultaneously improving productivity. Existing solutions are often expensive and inflexible. This thesis investigates how a pick- and-place routine, including depth camera data, can be integrated into a future modular and automated construction system that aims to provide a more cost- effective, automated solution by using off the shelf components. Initially, methods for identifying and picking construction material within the system are examined. To utilize the developed methods, a dedicated lifting tool and camera housing are designed. A system enabling control and feedback of the pick-and-place routine via an HMI is then developed. The resulting system is then evaluated through a series of tests where the results are compared to an ideal execution. The outcome of the work is a system that can be remotely operated, with a localization method that, when combined with the picking method, achieves a deviation of 2,86mm from the ideal placement. Finally, suggestions for improving the stability and reliability of the system is discussed.