Workload and production assessment in the avocado harvesting process using human-robot collaborative strategies

Human-robot collaboration strategies for agricultural applications are still a challenging research topic that is in its early stages of development. The analysis of the production and physical workload measures during the harvesting process can be useful to obtain relevant information to improve production and farm management. In this work, it is presented the assessment of the avocado harvesting process under different scenarios. For this purpose, two simulation environments were created and compared based on the obtained information from a Chilean farm. The first scenario uses only a human-based workforce to perform all the avocado harvesting-related tasks, whilst the second scenario is based on a human-robot cooperative strategy proposed to improve the harvesting procedure. A total of 41 different tests for each simulated scenario was performed, and then a comparison of them in terms of production and expected human physical workload was obtained. The results showed that the use of a cooperative strategy during harvesting tasks can increase the production measured from the cargo zone from 15% to 80%, as well as the total harvested production from 23% to 85%, whilst the human workload only increases from 1% to 16%. These results enclose the possibility of using production and workload measures as metrics for a human–machine interaction cooperative approach design, in benefit of the harvesting process management and decision making.