Submitted to: Irrigation Associations Exposition and Technical Conference Proceedings
Publication Type: Proceedings
Publication Acceptance Date: November 15, 2007
Publication Date: December 9, 2007
Citation: Oshaughnessy, S.A., Evett, S.R. 2007. IRT wireless interfact for automatic irrigation scheduling of a center pivot system. In: Proceedings of the 28th Annual International Irrigation Show, December 9-11, 2007, San Diego, California. p. 176-186. 2007 CDROM. Interpretive Summary: Growers are able to reduce their farming expenses by using sprinkler irrigation machines to decrease labor costs and improve water savings. Many modern center pivots are computer controlled and allow the grower to program a watering schedule. Sensor feedback provided to the center pivot computer, such as canopy temperature or soil moisture content, can improve irrigation scheduling by irrigating only when the crop needs water. This advanced automation will improve water conservation and water use efficiency. The optimal design for these sensors would include wireless data transfer and low power use. A low cost wireless sensor module was developed to work with an infrared thermometer. Two sets of eight wireless sensors were deployed into the field, each functioned as a separate system. One wireless sensor network (WSN), the Pivot WSN, was placed on the pivot span and the other, the Field-WSN, was placed in the crop. Experimentation over the cropping season showed that data from the sensor modules was transmitted more reliably from the Field WSN. We hypothesize that the metal structure of the center pivot interfered with the data transmission from the Pivot WSN. In the future, it very likely that sensor network systems will be combined with a commercial irrigation sprinkler to provide a more efficient irrigation system.
Technical Abstract: Infrared thermometers (IRTs) have been widely used in agricultural research as a method to measure canopy temperatures, an indicator of crop water stress. Although IRTs have proven to be reliable within the critical range for plant stress, they would be cumbersome for the grower to set up, maintain, and dismantle each irrigation season in a commercial system. A wireless sensor network of IRTs integrated into a center pivot lateral can facilitate the implementation of a fully automated irrigation system with sensors that can easily be mounted and dismounted from the system lateral line. The objectives of this study were to build an economical wireless interface for IRTs using radio frequency (RF) mesh networking modules and to investigate the network characteristics in a field application comparing mesh networking and simpler point-to-point networking. Our main hypothesis was that the mesh networking system was best suited for installation on the pivot lateral and its self-healing capabilities would overcome the majority of interference issues associated with the pivot's metal trusses, pipeline, and towers. The mesh networking architecture was expected to outperform the non-mesh network. Relatively inexpensive integrated silicon circuit components were utilized to construct the sensor interface module; the approximate cost was $150, which included the signal conditioning electronic circuit that interfaced the IRT with the microprocessor and the RF module, the battery, and the solar panel. As part of the network testing, the received signal strength index (RSSI) for two different antenna types was tested at two different heights above grade under the pivot and at thirteen different distances from the pivot point. The RSSI using a whip antenna was superior to that of a dipole antenna. Wireless sensor networks were deployed in the field (Field-WSN) and along the pivot lateral (Pivot-WSN) in point-to-point topologies using both non-mesh and mesh firmware, respectively. The Field-WSN outperformed the Pivot-WSN. Data packet retrieval was more than 90% successful for 93% of the growing season using the non-mesh networking firmware for the WSN established in the field crop. The Pivot-WSN data packet retrieval was more than 90% successful for 70% of the time using mesh-networking firmware, but data packet retrieval dropped significantly to < 80% success for 100% of the time when the firmware was changed to a non-mesh networking protocol during a trial period after the growing season. These results indicate the potential role of mesh networking and wireless sensors in agricultural field settings.