Artificial Intelligence for robotic sensing in industrial environment

Modern production facilities and infrastructure contribute to unnecessary waste of resources in several ways and thereby contributes to greenhouse gas emissions. The main challenge that exists today is how to economically detect all the different “sources of waste”. The typical production facility is very complex, hosts several thousands of different components (e.g. machines) as well as maintains a complex infrastructure to run the necessary production processes (e.g. steam, compressed air). The project aims to significantly reduce the effort to identify the sources of waste by fully automating the detection of “sources of waste” in facilities. This is done by using a modern and highly mobile robot (Spot, Boston Dynamics) which will be equipped with multiple sensors (e.g. ultrasound sensor, infrared camera, microphone, gas sensor). Based on a multisensory and sensor-fusioning set up, the focus of the project will be to develop an Artificial Intelligence which can fully automatically identify “sources of waste” to provide the maintenance department with the necessary basis of decision-making to stop the unnecessary waste of valuable resources in real-time and 24/7.