Multi-Modal Perception Using Vision-Based Tactile Sensors

Abstract

In agricultural technology, this thesis introduces a pioneering approach that leverages deep learning techniques for enhancing produce quality assessment, with a particular focus on exploring maturity and freshness in various fruits and vegetables. The central aim of this research is to innovate in the field of agricultural sensing by integrating vision and tactile sensing technologies, a step forward in the quest for more accurate and efficient agricultural practices. Furthermore, the thesis encompasses a comprehensive survey of existing vision-based tactile sensors, delving into their applications, datasets, and limitations. This survey not only provides a critical overview of the current state of the field but also sets the stage for future advancements. The thesis will explore the potential of multimodal data, utilizing RGB and tactile data, to enhance produce quality assessment. This innovative approach aims to compare the efficacy of multimodal sensing against traditional RGB-only or tactile-only methods. The objective is to establish a new benchmark in agricultural technology, offering insights into the most effective techniques for assessing produce maturity and freshness. Overall, this thesis aims to make a substantial contribution to the field of agricultural technology, bridging the gap between advanced sensing methodologies and practical agricultural applications, ultimately leading to more sustainable and efficient farming practices.

Date of Award	27 Jul 2024
Original language	American English
Supervisor	Naoufel Werghi (Supervisor)