Yukimasa Kaneda

A Reliable Wireless Control System for Tomato Hydroponics

Agricultural systems using advanced information and communication (ICT) technology can produce high-quality crops in a stable environment while decreasing the need for manual labor. The system collects a wide variety of environmental data and provides the precise cultivation control needed to produce high value-added crops; however, there are the problems of packet transmission errors in wireless sensor networks or system failure due to having the equipment in a hot and humid environment. In this paper, we propose a reliable wireless control system for hydroponic tomato cultivation using the 400 MHz wireless band and the IEEE 802.15.6 standard. The 400 MHz band, which is lower than the 2.4 GHz band, has good obstacle diffraction, and zero-data-loss communication is realized using the guaranteed time-slot method supported by the IEEE 802.15.6 standard. In addition, this system has fault tolerance and a self-healing function to recover from faults such as packet transmission failures due to deterioration of the wireless communication quality. In our basic experiments, the 400 MHz band wireless communication was not affected by the plants’ growth, and the packet error rate was less than that of the 2.4 GHz band. In summary, we achieved a real-time hydroponic liquid supply control with no data loss by applying a 400 MHz band WSN to hydroponic tomato cultivation.

Sliding window-based support vector regression for predicting micrometeorological data

A new methodology for predicting micrometeorological data is proposed.Our proposed method involves a novel combination of SVR and ensemble learning.Weak learners built from efficient extracted data is aggregated dynamically.Large-scale micrometeorological data to compare other methods is used.The best prediction performance and the lowest time complexity are achieved. Sensor network technology is becoming more widespread and sophisticated, and devices with many sensors, such as smartphones and sensor nodes, have been used extensively. Since these devices have more easily accumulated various kinds of micrometeorological data, such as temperature, humidity, and wind speed, an enormous amount of micrometeorological data has been accumulated. In recent years, it has been expected that such an enormous amount of data, called big data, will produce novel knowledge and value. Accordingly, many current applications have used data mining technology or machine learning to exploit big data. However, micrometeorological data has a complicated correlation among different features, and its characteristics change variously with time. Therefore, it is difficult to predict micrometeorological data accurately with low computational complexity even if state-of-the-art machine learning algorithms are used. In this paper, we propose a new methodology for predicting micrometeorological data, sliding window-based support vector regression (SW-SVR) that involves a novel combination of support vector regression (SVR) and ensemble learning. To represent complicated micrometeorological data easily, SW-SVR builds several SVRs specialized for each representative data group in various natural environments, such as different seasons and climates, and changes weights to aggregate the SVRs dynamically depending on the characteristics of test data. In our experiment, we predicted the temperature after 1h and 6 h by using large-scale micrometeorological data in Tokyo. As a result, regardless of testing periods, training periods, and prediction horizons, the prediction performance of SW-SVR was always greater than or equal to other general methods such as SVR, random forest, and gradient boosting. At the same time, SW-SVR reduced the building time remarkably compared with those of complicated models that have high prediction performance.

Proposal to Sliding Window-based Support Vector Regression☆

Abstract This paper proposes a new methodology, Sliding Window-based Support Vector Regression (SW-SVR), for micrometeorological data prediction. SVR is derived from a statistical learning theory and can be used to predict a quantity forward in time based on training that uses past data. Although SVR is superior to traditional learning algorithms such as Artificial Neural Network (ANN), it is difficult to choose the suitable amount of training data to build an optimum SVR model for micrometeorological data prediction. This paper revealed the periodic characteristics of micrometeorological data and evaluated SW-SVR can adapt the appropriate amount of training data to build an optimum SVR model automatically using parallel distributed processing. The future prediction experiment was conducted on air temperature of Sapporo, Tokyo, Hamamatsu, and Naha. As a result, SW-SVR has improved prediction accuracy in Sapporo, and Tokyo. In addition, it has reduced calculation time by more than 96% in all regions.

Greenhouse Environmental Control System Based on SW-SVR

Abstract Greenhouse environmental control systems using sensor networks are becoming more widespread and sophisticated. To match the produce of expert farmers, these systems collect data about cultivation environment and growth situation, and aim to control the environment for cultivating high quality crops. However, with no agriculture experience, it is difficult for system users to set control parameters of several devices properly. In order to reproduce prediction control performed by expert farmers’ cultivation without human intervention, the authors propose a smart greenhouse environmental control system based on sliding window-based support vector regression (SW-SVR). The proposed system performs prediction control based on accurate predictions in real time. SW-SVR is a new machine learning algorithm for time series data prediction. The prediction model automatically adjusts to the current environment periodically, predicts time series data with high accuracy and low computational complexity. The proposed system using SW-SVR enables system users to optimize controls for crops. Meanwhile, since plant growth is related to the photosynthesis and transpiration of leaves, the authors developed wireless scattered light sensors which measure leaf area size indirectly so as to estimate plant growth. Our experimental results, using data of scattered light sensors on-site, outside weather data, and forecast data as independent variables of SW-SVR for hydroponic culture of tomatoes, show the proposed system reduced prediction error of nitrogen absorption amount by 59.44% as Mean Absolute Error (MAE) and 52.89% as Root Mean Squared Error (RMSE) compared with SVR, and reduced training data by 43.07% on average. Furthermore, the sugar content of tomatoes cultivated by the prototype system increased 1.54 times compared with usual tomatoes.

Highly reliable wireless environmental control system for home gardening

Various agricultural environment control systems using sensor data have recently been proposed and implemented. However, the high temperatures and humidity in agricultural environments prevent wireless sensor networks from collecting environmental data and operating long term with high reliability. Therefore, in this paper, we propose a greenhouse environment control system that can maintain a high operating ratio and a low packet error rate (PER) even when operating in severe environmental conditions. Implementing a fault-tolerant system using 429MHz band wireless communication achieved an operating ratio of 100% and an average PER of 0.016%.

Multi-modal sliding window-based support vector regression for predicting plant water stress

Abstract Information communication technology (ICT) is required in the field of agriculture to solve problems arising because of the aging of farmers and shortage of heirs. In particular, environmental sensors and cameras are widely used in existing agricultural support systems for easy data collection. Although the traditional purpose of these systems is naive monitoring and controlling of the environment, the propagation of advanced cultivation is now expected by applying the data to machine learning and data mining technologies. Therefore, we propose a novel multi-modal sliding window-based support vector regression (multi-modal SW-SVR) method for accurate prediction of complicated water stress, which is a plant status, from two data types, namely environmental and plant image data. The proposed method includes two methodologies, SW-SVR and deep neural network (DNN) as a multi-modal feature extractor for SW-SVR. SW-SVR, which we proposed previously, is a suitable learning method for data with time-dependent characteristics, such as plant status. Moreover, we propose a new image feature, remarkable moving objects detected by adjacent optical flow (ROAF), to enable DNN to extract essential features easily for predicting water stress. Compared with existing regression models and features, the proposed multi-modal SW-SVR with ROAF demonstrates more precise and stable water stress prediction.

Motion-Specialized Deep Convolutional Descriptor for Plant Water Stress Estimation

Mechanical water stress assessment is needed in agriculture to mechanically cultivate high-sugar-content crops. Although previous methods estimate water stress accurately, no method has been practically applied yet due to the high cost of equipment. Thus, the previous methods have a trade-off relationship between cost and estimation accuracy. In this paper, we propose a method for estimating water stress on the basis of plant images and sensor data collected from inexpensive equipment. Specifically, a motion-specialized deep convolutional descriptor (MDCD), which is a novel image descriptor that extracts motion features among multiple sequential images without considering appearance in each image, expresses plant wilt strongly related to water stress. Implicit exclusion of appearance enables extraction of general features of plant wilt, which is insulated from the effect of differences in shapes and colors of places and individual plants. We evaluated the performance of the proposed method using enormous agricultural data collected from a greenhouse. Accordingly, the proposed method reduced the error of mean absolute error (MAE) by approximately 25% compared with a naive convolutional neural network (CNN) using original images. The results show that the MDCD enhances temporal information, while reducing spatial information, and expresses the features of plant wilt appropriately.

Highly reliable wireless control system for tomato nutriculture

In the field of horticulture, various environmental control systems have recently been proposed and implemented. Some systems target cultivation of high quality vegetables and require reliable environmental and device control data reception such as nutrient solution control for tomatoes. However, leaves or metal pipes prevent wireless sensor networks (WSNs) from collecting environmental data or controlling devices. Therefore, in this paper, we propose a wireless control system for tomato nutriculture even in severe conditions. Implementing a highly reliable wireless control system using 429MHz band and two channel access mechanisms, we archived wireless communication without packet loss.