Robot demonstration for horticulture



Over the last 8 years a joint venture between VHS and KMS Projects has made extensive trials with different camera / vision systems on different crops. The eventual aim of this is to develop a robotic harvester. After trying the different camera systems it was eventually decided that the 3D camera system would be the best for broccoli, to measure the size of each head and if they are within the required specification. 
To date a vision guidance system to control a robotic arm has been designed. The research so far has been accomplished with no grants and very limited funds. To be able to demonstrate to the horticultural industry the potential funding is sought to rent the key components, manufacture the required toolbar and fit it all on a tractor to demonstrate how a selective robot could work in a realistic practical situation.   

Benefits to Industry:

The labour cost for manual handling of ornamentals and harvesting of edible crops is a large proportion of total costs in many horticultural businesses. Potential labour shortage or unavailability is a major risk factor for many businesses. Automation of the processes would reduce risk and cost and make businesses more competitive.

The benefits go beyond the saving of labour costs. For example if processes are automated they can take place at night when temperatures are lower, thereby reducing refrigeration costs and improving shelf life and product quality.

The Brassica Growers Association has identified development of an automated harvester as one of their top priorities. By inviting selected growers from all sectors of horticulture to a demonstration the potential of the technology to develop for different uses can be maximised.

Project code:
CP 153
15 June 2015 - 15 October 2015
AHDB Horticulture
AHDB sector cost:
Project leader:


CP 153_KT Summary

About this project

Aims and Objective:

To replace manual operations in horticulture with automated processes that can replace the hand/eye co-ordination of a human operator.

Initial objective

  • To demonstrate the integrated use of images from a 3D camera enabling a robotic arm to respond to specified crop dimensions. The vision guidance system will be demonstrated by its ability to detect the size and shape of broccoli heads and control a robot to select heads that fall within specified dimensions.