2019-01507 - Post-Doctoral Research Visit F/M Energy efficient wireless data acquisition and prediction system for smart agriculture

Contract type : Public service fixed-term contract

Level of qualifications required : PhD or equivalent

Fonction : Post-Doctoral Research Visit

Context

This position will run in the framework of the LIRIMA http://www.lirima.uninet.cm AgriNet project.

The AgriNet https://team.inria.fr/agrinet/ project, a joint project between Inria and Stellenbosch University, aims to design and deploy a full wireless sensor network based system to sense and collect crop data, to assist with potato and vineyard production. The system will consist of a smart wireless sensor network to sense and efficiently collect crop and environmental data to feed into a predictive and decision making tool. 

Assignment

Energy efficient wireless data acquisition and prediction system for smart agriculture

-----------------------

The AgriNet project, a joint project between Inria and Stellenbosch University, aims to design and deploy a full wireless sensor network based system to sense and collect crop data, to assist with potato and vineyard production. The system will consist of a smart wireless sensor network to sense and efficiently collect crop and environmental data to feed into a predictive and decision making tool. 

 

The main goal of this investigation will be the design of a predictive model dedicated to agro-parameters, particularly focused on potatoes. The model will utilize the distributed information obtained from the sensing network and should be able to recognize and predict key crop parameters and tendencies. This will be based on a two-fold approach: Local predictive algorithms will allow the reduction of the amount of data to be sent over the network and thus the reduction of energy consumption. Secondly, the model will feed a cloud based predictive tool to assist producers to better utilise their resources. In the first case, the local prediction will be computationally light, since only a small set of data is involved.

Machine learning based models will be used and applied to our specific case. For the local version, techniques will be adapted to take the constrained hardware resources of devices into account.

This work will be conducted jointly with the Inria FUN group and two departments of the Stellenbosch University. These partners provide very strong combined expertise in networking, data analysis, machine learning and agricultural knowledge.  

The results of this work will then be implemented and run in a trial.

Main activities

  • Design of machine learning based predictive tools for smart agriculture
  • Conduct a survey of similar existing techniques
  • Implementation of the designed solution

Skills

Skills

  • Knowledge and experience of wireless networks and/or edge/fog architecture
  • Proven skills in simulation tools and development

 

Qualities

  • English speaking
  • Autonomy
  • Unbiased research view
  • Team based approach
  • Capacity to write English reports and papers

 

Benefits package

  • Subsidized meals
  • Partial reimbursement of public transport costs
  • Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
  • Possibility of teleworking (after 6 months of employment) and flexible organization of working hours
  • Professional equipment available (videoconferencing, loan of computer equipment, etc.)
  • Social, cultural and sports events and activities
  • Access to vocational training
  • Social security coverage

Remuneration

Net monthly salary (after taxes) : 2132.97€