The world’s demand for agricultural products is growing rapidly requiring an estimated 50% increase in agricultural productivity in the next 30 years. There is a strong need for more sustainable agriculture to lower the impact on the environment. Agriculture furthermore suffers from a lack of skilled labour. Agricultural robotics and precision farming can over a part of the solution to meet these challenges.
Challenges for agricultural robotics
There are three challenges for robots to operate in agri-food environments: (1) the variation in the appearance of objects, environmental properties, cultivation systems and tasks, (2) incomplete information due to occlusions, sensor noise and uncertainty, and (3) safety in the interaction with fragile plants and produce, and human beings.
To tackle these challenges, the research in my group targets at the following topics:
- Robust perception. Deep neural networks have revolutionised the field of machine vision. Compared to traditional image-processing algorithms, deep neural nets can deal better with variation in the appearance of objects and differences in illumination conditions. To further improve the performance, we study the generalisability of neural networks and develop methods to deal better with variations.
- Active perception. To deal with occlusions, robots need to actively perceive the environment. We study methods for active perception, to allow robots to decide on new viewpoints to acquire relevant information from the environment to perform its tasks.
- Soft robotics. To safely interact with the agri-food environment, robots need to get a soft touch. On the one hand, this requires the use of soft material and actuators. On the other hand, it requires tactile sensors and control algorithms.