This paper presents a new form of piecewise-affine (PWA) solution, referred to as PWA hierarchical (PWAH), to approximate the explicit model predictive control (MPC) law, achieving a very rapid control response with the use of very few computational and memory resources. This is possible because PWAH controllers consist of single-input single-output PWA modules connected in cascade so that the parameters needed to define them increase linearly instead of exponentially with the input dimension of the control problem. PWAH controllers are not universal approximators but several explicit MPC controllers can be efficiently approximated by them. A methodology to design PWAH controllers is presented and validated with application examples already solved by MPC approaches. The designed PWAH controllers implemented in field-programmable gate arrays provide the highest control speed using the fewest resources compared with the other digital implementations reported in the literature.
Precision agriculture can be defined as the science of using technology to improve the agricultural production. It is advisable for farmers to use a decision support system; in fact, real–time monitoring of climatic conditions is the only way to know the water needed by a cultivation. On the other hand, since the 1990s, a strong decrease of the Mediterranean Quercus has been observed in the pastures of southwestern Spain and Portugal, causing a high mortality of holm and cork oak trees. Among the factors associated with this decrease, the radical decomposition caused by Phytophthora Cinnamomi is remarkable for its gravity, which makes it necessary to reforest the trees and to monitor the microclimatic factors that have an influence on this regeneration. Wireless Sensor Networks (WSN) are a technology in full evolution and development, as well as their appropriate use in cultivations that help to implement ecological techniques. With these considerations in this work five units/nodes with one or more sensors that allow different environmental readings have been developed. In this work, the acquisition of data obtained from different sensors has been achieved, allowing the monitoring of climatic elements such as soil moisture, air quality, temperature and humidity, rainfall intensity, precipitation level, wind speed and direction, luminous flux and atmospheric pressure. A web page has been designed where the user can consult the climatic conditions of the cultivation or reforestation. Different devices interconnected with a central unit have been developed where measurements of the cultivation are sent for its later analysis by the farmer. The microclimatic data acquisition developed in the WSN proposed in this paper allows a farmer to make decisions about the irrigation of the cultivation, use of fertilizers, the development and maturation phases of the cultivated products, obtaining the optimum stages of cultivation and harvesting.
Since 1992, Xfuzzy environment has been improving to ease the design of fuzzy systems. The current version, Xfuzzy 3, which is entirely programmed in Java, includes a wide set of new featured tools that allow automating the whole design process of a fuzzy logic based system: from its description (in the XFL3 language) to its synthesis in C, C++ or Java (to be included in software projects) or in VHDL (for hardware projects). The new features of the current version have been exploited in different application areas such as autonomous robot navigation and image processing.
This paper proposes the design of hierarchical piecewise-affine (PWA) controllers to alleviate the processing time or prohibitive memory requirements of large controller structures. The constituent PWA modules of the hierarchical solution have fewer inputs and/or coarser partitions, so that they can reduce considerably the hardware resources required and/or the time response of the controller. A design methodology aided by CAD tools is employed to design the parameters of the controller, implement its architecture in an FPGA, and verify the static and dynamic behavior of the digital implementation by applying hardware-in-the-loop testing.