Currently, the detection of soil pollution is done by manual surveys which are inefficient, unreliable, time consuming and very expensive. The need of a better and low-cost technical approach, that would be able to assess contaminated areas before programing the required remediation actions, has become more intense due to public awareness for the environment. This project aims to develop a semi-autonomous, remotely controlled robotic system that will deploy three novel (and a modified existing) sensors for the assessment of soil quality and site characterization. The large contaminated sites will be covered by 100%, presented by fine scale 3D maps of contaminant levels and giving quality and quantity data for the pollutants (heavy metals, organic solvents, NAPLS) without the need of human intervention. Considering that the personnel is endangered by the toxic contaminants, this is very important.

Four different types of sensors (HPU, GPR, chemical sensors and SAW) will be integrated to detect efficiently the contaminants in the surface and underground (up to 10 meters). Data collected from the sensors will be fused, correlated and optimized to provide GPS linked ground 3D images of contaminant distribution. Thus ground property images from different sensors can be combined to produce composite images that can be used also later to during the treatment of soil contamination. Advanced signal processing techniques will be used to obtain the correct interpretation of the data acquired by the sensors and image processing will ameliorate the fused images. The automated inspection soil contamination system will benefit the SMEs involved in NDT, robotics and contamination detection.