Affordable remote monitoring of plant growth in facilities using Raspberry Pi computers

A low-cost, mobile real-time kinematic geolocation service for engineering and research applications

Centimetre accurate geolocation service is beneficial to a wide range of applications, ranging from sports engineering, civil infrastructure, autonomous vehicles, surveying to digitisation of historically significant structures. Previously, these features were confined to prohibitively expensive commercial hardware, requiring technical knowledge and experience to operate. Continued technological advancements have seen the miniaturisation of electronics and antennas, coupled with an increase in the number and performance of global navigation satellite systems (GNSS) by various nations and organisations, providing global signal coverage. This paper demonstrates a low-cost, mobile, real-time kinematic (RTK) geolocation service for engineering and research applications, fabricated from components readily available from commercial suppliers. This solution, consisting of a mobile RTK base station and RTK rover, provides centimetre-accuracy performance up to a distance of 15 km away from the base station. Correction data is transmitted over the internet using free and open software solutions. The small footprint of both the RTK base station and RTK rover, provides versatile applications even in remote locations. The performance of the geolocation service is validated using field experiments, comparing measurements against state-of-the-art photogrammetry, light detection and ranging (LiDAR) and digital level measurement technologies. The authors encourage the adoption of the RTK geolocation solution based on the calibrated results.

smAvo and smaTo: A fruity odyssey of smart sensor platforms in Southern Africa

Avocado and tomato production are key agricultural sectors for many economies including South Africa. The quality of avocadoes and tomatoes that reach consumers is directly dependent on the handling at the postharvest stage. The fruit undergoes an extensive journey, and subsequently stress, comprised of several steps, including harvesting, application of postharvest treatments, packaging and transportation by road and sea to reach the export markets. To date, the quantification of these external stresses on a discrete fruit has not been measured in detail, primarily due to a lack of compatible instrumentation. This paper demonstrates a low cost, stand-alone, open source data acquisition system, termed smAvo and smaTo, that can be introduced effortlessly to monitor agricultural processing facilities and transportation networks. The miniaturised Arduino sensor platform is enclosed within a waterproof enclosure and surrounded by a 3D printed shell manufactured from morphologically compatible materials. The software is customisable to the needs of the research project or individual transportation phases under study, providing both extensive environmental data and high-frequency tri-axis acceleration measurements that are crucial to understanding the dynamic processes that directly affect the final quality of the fruit. Four successful field trials demonstrate the fidelity of both the smAvo and smaTo platforms.

Fly-by-Pi: Open source closed-loop control for geotechnical centrifuge testing applications

Geotechnical centrifuges are valuable instruments for physical modelling of complex geotechnical problems in a controlled laboratory setting. In comparison to full-scale testing, scaled models are cost effective to construct and instrument and, when tested in a geotechnical centrifuge at increased centrifugal accelerations, are capable of replicating full-scale stress-strain soil behaviour. Centrifuge modellers require specialised hardware and instruments capable of functioning under high accelerations. Such hardware is costly, nearly always purpose built, and often rely on commercial, closed-source data acquisition systems, hardware and control systems. This paper demonstrates a novel and versatile, low cost, open source logger and control system that works in parallel alongside existing centrifuge hardware. This solution, termed Fly-by-Pi, was developed using the Raspberry Pi microcomputer. The system provides closed-loop control of linear actuators with the ability to operate in either cyclic, monotonic, or static load- or displacement-control. The control mechanism can be reprogrammed according to experimental requirements, even during flight in the centrifuge. Three independent experiments are described which included the Fly-by-Pi controller as a key component in their operation. Based on the experience gained during these experiments, the authors encourage wide-spread adoption of open-sourced hardware solutions in extreme testing environments.

iFO (infrared Fish Observation) - An open source low-cost infrared underwater video system

Underwater video surveillance is an important data source in marine science, e.g. for behaviour studies. Scientists commonly use water resistant ruggedized monitoring equipment, which is cost-intensive and usually limited to visible light. This has two disadvantages: the observation is limited to space and time where visible light is available or, under artificial illumination, behaviour of marine life is potentially biased. Infrared (IR) video surveillance have been used before to overcome these. It records videos at visible light and under IR-illumination. With today's efficiency of IR-LED and video technology even low-cost systems reach visibility ranges suited for many application scenarios. We describe a low-cost open-source based hardware/software system (iFO). It consists of a single-board computer controlling the camera and lamps (with high power IR-LEDs), printed circuit boards (PCB), the underwater housings and 3D-printable models to mount PCBs in the housings and the housings to standard GoPro mounts. The Linux based software includes webserver, remote control, motion detection, scheduler, video transfer, storage at external hard disk and more. A ready-to-use SD-card image is included. We use rugged underwater housings with 100 m (optional 400 m) depth ratings. Finally, we describe a typical application observing the behaviour of cod in fish pots.

Greenotyper: Image-Based Plant Phenotyping Using Distributed Computing and Deep Learning

Image-based phenotype data with high temporal resolution offers advantages over end-point measurements in plant quantitative genetics experiments, because growth dynamics can be assessed and analysed for genotype-phenotype association. Recently, network-based camera systems have been deployed as customizable, low-cost phenotyping solutions. Here, we implemented a large, automated image-capture system based on distributed computing using 180 networked Raspberry Pi units that could simultaneously monitor 1,800 white clover (Trifolium repens) plants. The camera system proved stable with an average uptime of 96% across all 180 cameras. For analysis of the captured images, we developed the Greenotyper image analysis pipeline. It detected the location of the plants with a bounding box accuracy of 97.98%, and the U-net-based plant segmentation had an intersection over union accuracy of 0.84 and a pixel accuracy of 0.95. We used Greenotyper to analyze a total of 355,027 images, which required 24-36 h. Automated phenotyping using a large number of static cameras and plants thus proved a cost-effective alternative to systems relying on conveyor belts or mobile cameras.