Robotic Perception in Agri-food Manipulation: A Review

TL;DR

This paper reviews modern robotic perception models and their effectiveness in agricultural robotics, focusing on scene understanding and object detection crucial for tasks like phenotyping, quality analysis, and harvesting.

Contribution

It provides a comprehensive overview of current perception models and evaluates their applicability to agri-food manipulation tasks.

Findings

Perception models vary in effectiveness for agricultural tasks

Scene understanding is essential for autonomous harvesting

Robotic perception models need domain-specific adaptation

Abstract

To better optimise the global food supply chain, robotic solutions are needed to automate tasks currently completed by humans. Namely, phenotyping, quality analysis and harvesting are all open problems in the field of agricultural robotics. Robotic perception is a key challenge for autonomous solutions to such problems as scene understanding and object detection are vital prerequisites to any grasping tasks that a robot may undertake. This work conducts a brief review of modern robot perception models and discusses their efficacy within the agri-food domain.