In the heart of Portugal, researchers are tackling one of agriculture’s most pressing issues: labor shortages and the high cost of automation. Vítor Tinoco, a researcher at INESC TEC—Institute for Systems and Computer Engineering, Technology and Science in Porto, has been leading a team that’s developing low-cost robotic manipulators to revolutionize agricultural practices. Their latest work, published in the journal Sensors, benchmarks three control strategies for these affordable machines, paving the way for more efficient and cost-effective agricultural automation.
The global population is projected to reach 9.7 billion by 2050, driving up the demand for agricultural products. However, labor shortages, especially during harvest seasons, are a significant challenge. “Engaging in harvesting demands extensive labor hours and proves physically taxing and repetitive,” Tinoco explains. “Moreover, these tasks are typically seasonal, dissuading workers from risking unemployment until the next season.” This is where agricultural robots come in, taking over repetitive tasks and reducing the physical strain on human workers.
But here’s the catch: commercially available robotic manipulators are expensive, ranging from EUR 3,000 to EUR 500,000. This is where Tinoco’s research comes in. His team has been developing and benchmarking three control strategies for low-cost Selective Compliance Assembly Robot Arm (SCARA) manipulators: a Sliding Mode Controller (SMC), a Reinforcement Learning (RL) Controller, and a novel Proportional-Integral (PI) controller with a self-tuning feedforward element (PIFF).
Each controller has its strengths and weaknesses. The SMC, for instance, offers the best response time but comes with joint movement jitter, which could potentially damage crops. The RL controller, on the other hand, shows sudden breaks and overshoots upon reaching the setpoint. The PIFF controller, however, provides the smoothest reference tracking but is more susceptible to changes in system dynamics.
So, how might this research shape future developments in the field? Tinoco believes that integrating features from state-of-the-art control strategies could lead to more effective control solutions for low-cost agricultural SCARA manipulators. “Future research should focus on hybrid approaches that combine these strengths to address the specific challenges of agricultural automation,” he suggests.
The implications for the agricultural sector are significant. By making robotic manipulators more affordable and efficient, this research could help address labor shortages and increase agricultural productivity. It could also pave the way for more sustainable farming practices, as robots can work around the clock without the need for breaks, reducing the environmental impact of agriculture.
Moreover, the energy sector could also benefit from this research. As the demand for agricultural products increases, so does the need for energy to power the machinery used in farming. More efficient robotic manipulators could help reduce energy consumption, making agriculture more sustainable and cost-effective.
The research, published in the journal Sensors (translated from the Portuguese title ‘Sensores’), is a significant step forward in the field of agricultural automation. It provides a benchmark for future research and development in this area, paving the way for more efficient and cost-effective agricultural practices. As Tinoco and his team continue their work, the future of agriculture looks increasingly automated and sustainable.