Integrate machine learning techniques to improve detection/discrimination.

The candidate MUST speak Spanish and English.

Task 1: Integrate machine learning techniques to improve detection/discrimination. Cloud computing platform will be designed to store the collected data from the sensor head. It will allow to get data snapshots for training AI as well as for analytics. Also the data collection and labelling process will be deployed. Finally, the AI will be trained and deployed onto the mobile hardware to detect weeds in the real time.

The following activities will be done:

• Upload and integrate weather forecasts from the European Centre for Medium Range Weather Forecasting (ECMWF) and their interconnection through the development of Big Data solutions based on the construction of predictive Deep Learning models for both the growth and health of crops and for the presence and evolution of weeds and pests.

• Develop and code of a web interface for end-user private access to remotely visualise farm status, scanned by our sensor head; information about weed infestations, pests, crop status, etc. will be provided.

• Develop and code of an operator console, installed inside the vehicle, configurable and equipped with all the information related to the operating status of the robot, crop growth indicators, present and future heat maps of pests and weeds, irrigation and fertilization recommendations and alerts of incoming adverse weather conditions.

• Development of the AI algorithm; development, configuration and training (machine learning) of the neural network for recognition of the different species of weeds and pests. Increase up to +25% additional weeds and +30% pests from current status.