Going beyond convenience: Using IoT to create ‘intelligent’ irrigation systems
By Jiang Hu, PhD, and Charles Hillyer, PhD
Technology advances have brought numerous bells and whistles that can potentially improve water-use efficiency. However, there lacks a complete solution that makes use of all leverages and provides growers with peace of mind.
Nowadays, irrigation machines can feature site-specific variable rate irrigation, but it comes with limited management support on how to efficiently utilize such a feature. Wireless sensors indicate a certain degree of water need, and sensor-based irrigation does improve water savings. However, few sensor-based systems account for weather conditions or possible precipitation of next to a few days. Moreover, wireless sensors entail significant reliability issues, such as communication failure, battery depletion and sensor aging effects. The substantial maintenance requirements easily drive potential users away.
Using the internet to control & manage
The “internet of things,” or IoT, promises to bring a new way to control and manage irrigation. According to Wikipedia, the IoT is “the network of physical devices, vehicles, and other items embedded with electronics, software, sensors, actuators, and network connectivity which enable these objects to collect and exchange data.” Existing IoT-based irrigation systems allow
14 Irrigation TODAY | January 2018
growers to remotely monitor and control irrigation machines. However, without making a connection between sensors and controls, such systems are merely a means of convenience rather than an actual promise on water savings.
An intelligent solution
The market calls for a complete solution for intelligent and automatic variable rate irrigation. But what would such a system look like? A fully automatic system must integrate spatial data from multiple sensors, past water applications and weather forecasts. At the same time, the system must be tolerant of faults and breakdowns. One such prototype system is under development at Texas A&M and is near completion. It considers soil moisture measurements from multiple sensors, as well as forecasts of precipitation and evapotranspiration [ET]. The IoT, or the connection with the internet, facilitates automatic information collection for irrigation decision-making. This is quite different from prior methods where the internet merely provides a user interface.
The centerpiece, also the brain, of the system is an intelligent control algorithm (or mathematical procedure). The algorithm estimates the amount of water needed for each location according to current soil moisture level and considers the probability of near future precipitation and ET. The algorithm uses a “machine learning” procedure that is sensitive to crop and soil types. Machine learning is a form of artificial intelligence in which computers are “trained” to make human-like decisions. Examples of machine learning applications include facial recognition, language translation and self-driving cars.
The machine learning algorithm for an irrigation system considers multiple scenarios based on crop simulations and probabilistic estimates of future conditions. By considering all available information and using machine learning, the system achieves autonomous control with the minimum need of human intervention. However, the system permits a manual mode for scenarios where the user desires direct control.
Wireless sensor
NEW YEAR, NEW TECHNOLOGIES
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40
