Short name: IIoT-REPLAN
Long name: Industrial IoT-dRiven rEmote Path pLANning

Company: Queen’s University Belfast |
Country: United Kingdom

Call: F4Fp-04-M (see call details)
Proposal number: F4Fp-04-M08


Computation off loading in everyday real world applications is becoming increasingly important. In Industry 4.0 and especially in collaborative robotics where humans and robots need to work together in dynamic environments, it is becoming a necessity; large amount of information needs to be processed and complex algorithms must be executed on limited resources hardware. IIoT-REPLAN explores the offloading possibilities of the decision-making and monitoring tasks in path planning of autonomous agents. The image processing and decision algorithms, especially in time-varying and non convex environments, are computationally intensive and energy-hungry tasks. The proposed experiment will realize an IIoT-enabled assistive remote path planning mechanism and find the expected gains of computation off loading in the cloud, by using the edge computing and sensor facilities of FED4FIRE+testbeds, namely the NETMODE and Virtual Wall.The experiments are designed to expose the trade offs between computing and communication resources with a focus on in control design and implementation. The NETMODE testbed is ideal, as the wireless nodes are placed in a populated building that emulates the factory floor, i.e., a time-varying and non convex environment. The experiment utilizes Raspberry P is mounted on mobile robots and equipped with low-cost visual and motion sensors.

The project objectives are outlined below:

  1. Design and implementation of local and remote path planning controllers, and of the scheduling mechanism that decides their optimal interaction.
  2. Design and development of a computation offloading mechanism dynamically allocating the resources of the edge/cloud resources, co-designed with the control algorithm for the path planning problem.
  3. Exhaustive experimentation and impact exploration using well-defined metrics such as the accuracy of the produced trajectory, responsiveness, robustness, resilience, and energy consumption.In addition to the expected scientific impact, IIoT-REPLAN aspires to provide various types of feedback to the Fed4FIRE+ consortium, enhancing its functionality, while increasing its visibility and future potential.


  • Single Vision-Based Self-Localization for Autonomous Robotic Agents (link will be available soon)

Start typing and press Enter to search