Thursday, February 23, 2017
The Challenges for Artificial Intelligence in Agriculture.
The Challenges for Artificial Intelligence in Agriculture. A group of maize farmers stands huddled around an agronomist and his computer on the side of an irrigation pivot in central South Africa. The agronomist has just flown over the pivot with a hybrid UAV that takes off and lands using propellers yet maintains distance and speed for scanning vast hectares of land through the use of its fixed wings. The UAV is fitted with a four spectral band precision sensor that conducts onboard processing immediately after the flight, allowing farmers and field staff to address, almost immediately, any crop anomalies that the sensor may have recorded, making the data collection truly real-time. In this instance, the farmers and agronomist are looking to specialized software to give them an accurate plant population count. It’s been 10 days since the maize emerged and the farmer wants to determine if there are any parts of the field that require replanting due to a lack of emergence or wind damage, which can be severe in the early stages of the summer rainy season. At this growth stage of the plant’s development, the farmer has another 10 days to conduct any replanting before the majority of his fertilizer and chemical applications need to occur. Once these have been applied, it becomes economically unviable to take corrective action, making any further collected data historical and useful only to inform future practices for the season to come.more