Robotic Picking of Tangle-prone Materials (with Applications to Agriculture).

Student thesis: Doctoral ThesisDoctor of Philosophy


The picking of one or more objects from an unsorted pile continues to be non-trivial for robotic systems. This is especially so when the pile consists of individual items that tangle with one another, causing more to be picked out than desired. One of the key features of such tangling-prone materials (e.g., herbs, salads) is the presence of protrusions (e.g., leaves) extending out from the main body of items in the pile.

This thesis explores the issue of picking excess mass due to entanglement such as occurs in bins composed of tangling-prone materials (TPs), especially in the context of a one-shot mass-constrained robotic bin-picking task. Specifically, it proposes a human-inspired entanglement reduction method for making the picking of TPs more predictable. The primary approach is to directly counter entanglement through pile interaction with an aim of reducing it to a level where the picked mass is predictable, instead of avoiding entanglement by picking from collision or entanglement-free points or regions. Taking this perspective, several contributions are presented that (i) improve the understanding of the phenomenon of entanglement and (ii) reduce the picking error (PE) by effectively countering entanglement in a TP pile.

First, it studies the mechanics of a variety of TPs improving the understanding of the phenomenon of entanglement as observed in TP bins. It reports experiments with a real robot in which picking TPs with different protrusion lengths (PLs) results in up to a 76% increase in picked mass variance, suggesting PL be an informative feature in the design of picking strategies. Moreover, to counter the inherent entanglement in a TP pile, it proposes a new Spread-and-Pick (SnP) approach that significantly reduces entanglement, making picking more consistent. Compared to prior approaches that seek to pick from a tangle-free point in the pile, the proposed method results in a decrease in PE of up to 51% and shows good generalisation to previously unseen TPs.
Date of Award1 Aug 2023
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorMatthew Howard (Supervisor) & Simon Parsons (Supervisor)


  • Robotic bin-picking of tangling-prone materials
  • Entanglement reduction
  • Robotics in Agriculture and Forestry
  • Agricultural Automation
  • Computer Vision for Automation

Cite this