top of page

          The general objective of the project MultiMonD2 is to design, build and develop automated systems (robotic vectors), capable to offer support and mobility for micro-laboratories specialized for different analysis necessary to the study of the water quality.
          The vectors from the platform MultiMonD2 can be connected to the micro-laboratories with different functions, for the monitorization of the water quality on the Romanian sector of the Danube and the Danube Delta, and testing depollution capabilities. Due to the architecture of the communication system between the robotic vectors (drone / boat) equipped with sensors and video camera and a control center (3D VERO VIPRO MUTIMOND2), the micro-laboratories can work independently or can be used together in case of simultaneous interventions on land and water. The control station GCS mobile MMD2 of the system MultiMonD2 was designed to offer maximum performance. The Software for the execution of aerial, bathymetry and navigation maps can be used in design. The information received from the robotic systems is recorded for the purpose of analysis and further processing. The GCS is designed to be compact and light, in order to be transported by a human or mounted on the automobile that transport the robotic systems. By using a mobile internet connection, data received from the robotic systems and sensors can be transmitted to mobile or fixed control centers.
         Thus, the micro-laboratories of the platform based on robotic vectors were developed and tested:

  • micro-laboratory for detection and monitoring based on air vectors:

    1) drone equipped with sensors with an autonomy of approximately 30 minutes;

​       2) drone equipped with LIDAR system and surveillance video camera;

       3) VAM-T multirotor equipped with specific sensors for monitoring the course of the Danube;

  • micro-laboratory for detection and monitoring based on aquatic vectors: boat equipped with sensors and with the components of the electronic control structure, with an autonomy of approximately 2 hours;

  •  micro-laboratory for the local chemical decontamination based on a mobile plasma system that can be transported in different locations (height 115 cm, 80 cm x 60 cm width / depth and weight of 100 kg) and work anywhere without the need for additional cooling of plasma sources;

  • micro-laboratory designed with sensors for the detection of toxic gases

SnO2-based sensors for pollution monitoring can be attached to any robotic vector.​

         

          

1.jpg
3.png
4.jpg

FIG. A) MultiMonD2 platform based on robotic vectors that can be connected to microlaboratories with different functionalities, for monitoring water qualities on the Romanian sector of the Danube and Danube Delta and testing depollution capabilities: a) air vector (drone), b) air vector drone with LIDAR system), c) air vector (VAM-T multirotor), d) portable decontamination system and e) Aquatic vector (boat)

          Ground Control Station (GCS) module. The module for inference detection  (MID) developed to continue the tests, performs video processing, being integrated in the GCS control system as a module that works decentralized but also in cooperation with other modules, with remote communication by integration on drone or boat. The interdependence between MID and GCS and the sequence of data flow is shown in Figure B. The transmission and reception of data to the central unit of the GCS is done using the UDP communication protocol. The data transferred between the GCS and the Detection System are in JSON format. In addition, images or video streams can be transmitted from MID to GCS. By using the JSON format, the status of the LEDs on the GCS corresponding to the robot vector detections is communicated and modified. MID transmits to GCS the information related to the size of the objects, the class they belong to, the confidence score, the distance to the object, etc. This information is filtered by GCS and passed on to other modules.
 

10.jpg

Fig. B) Modul inferențial și server Deep Learning

bottom of page