BunkBot lot feeder demonstration site

Summary

With a view towards bunk reading autonomy, MLA project B.FLT.1006 allowed delivery of a campaign of experiments for a prototype autonomous mobile robot for the Bunk Scanner. The robot has been branded the BunkBot. This experiment demonstrated that under normal operating conditions the prototype BunkBot (complete with Bunk Scanner payload) predicted feed remaining in bunks accurately and repeatedly, outperforming human callers in both criteria. Most importantly, the higher performance bunk reads were performed without human input or intervention.

While this preceding research effort output a working prototype, it was not integrated into the day-to-day bunk calling operation of the feedlot.

Pinegrove Feedlot – owned by Mort & Co Lot Feeders – expressed interest in integrating the system into routine bunk management practices under both day and night conditions. Furthermore, Pinegrove offered to be a demonstration site for the feedlot industry, with planned monthly tours for industry stakeholders.

The demonstration site hosted a total of 44 lot feeders over five demonstration days. Key integration learnings were gained from the commissioning period at the demonstration site and the BunkBot is now operating autonomously at Pinegrove Feedlot.

Objectives

(1) Support activities to enable commercial lot feeders opportunities to view the BunkBot integrated into routine commercial bunk management at Mort & Co Lot Feeders Pinegrove Feedlot.

(2) Develop key integration learnings from the commissioning period.

Key findings

• A total of 44 lot feeders attended the demonstration site over five field days.
• During the demonstration period, the BunkBot travelled at an average speed of 7km/hr (range from 5 to 12km/hr).
• A new lithium ion battery array was installed on the BunkBot to enable recharging cycles.
• The BunkBot is now accomplishing three bunk calls per day at Pinegrove feedlot.

Benefits to industry

The BunkBot, when integrated with Bunk Scanner technology has the ability to obtain data from late night and early morning bunk reads. The scanner can also read bunks with greater precision and accuracy than humans, bringing consistency to the bunk management process. Further research will explore if this brings productivity gains to commercial feedlots.

MLA action

The BunkBot was demonstrated to feedlot stakeholders at five separate events.

Future research

Future research will explore the value of night time bunk reads on cattle performance. In addition, further research will focus on expanding bunk management algorithms for feeding cattle across diverse market categories. 

A parallel project (MLA Project B.FLT.1020) has delivered a safety management plan framework for sites considering implementation of autonomous vehicles.