Andre Riyuiti Hirakawa

Robotic mapping and navigation in unknown environments using adaptive automata

Real mobile robots should be able to build a virtual representation of the physical environment, in order to navigate and work in such environment. This paper presents an adaptive way to make such representation without any a priori information of the environment. The proposed system allows the robot to explore the entire environment and acquire the information incoming from the sensors while it travels and, due to the adaptability of the mapping method, the system is able to increase the memory usage according to the already mapped area. The map, built using the adaptive technique, is useful to provide navigation information for the robot, allowing it to move on the environment.

Robotics applied to work conditions improvement in power distribution lines maintenance

Based on the analysis of the tasks concerning maintenance of power distribution networks from the Ergonomics point of view, Paulista Power & Light Company (CPFL, Brazil) and the University of Sao Paulo (USP, Brazil) started a project focused on the application of Mechatronics and Robotics techniques and knowledge to improve the maintenance activity, also establishing a roadmap of developments to be conducted. As a pioneer research & development on that, four devices are proposed and described herein: (a) A computer-controlled automatic elevator attached to a pickup type lightweight vehicle that helps electricians to reach high positions replacing the conventional ladder; (b) A flexible telescopic robot for inspecting the wood cross arm from the ground; (c) A weight balancer to help electricians to execute tree trimming around distribution lines, by reducing the physical effort required for holding and manipulating the trimming equipment; (d) And a mechatronics version of the weight balancer able to automatically adapt to different loads.

Biologically-inspired Neural Network for Traffic Signal Control

Urban mobility is a central concern of large cities around the world. The growing urbanization indicates the situation can be even worse. A traffic demand higher than the urban capacity generates traffic congestions, which can be reduced through an efficient traffic signal control method. This paper presents a Biologically-Inspired Neural Network for traffic signal control. Instead of focusing on the macroscopic optimization of urban traffic, like other works, the hereby proposed control method investigates a single intersection between streets. This way, it is possible to incorporate more knowledge about the system dynamics into the control model and analyze its effects on control efficiency. The proposed method consists of a competitive neural network, which balances feedforward and feedback inhibition to synchronize the activity of the neurons, and, thus, the semaphore activation. Moreover, other proprieties of biological neurons are adopted: intrinsic plasticity, to impose system constraints; and synaptic plasticity, to prioritize traffic flows. The flexibility of the neurons and its synaptic connections regarding parameter definition constitute the capacity of easily incorporating knowledge about the system dynamics into the control model. Results of comparative simulations validate the proposed method and illustrate its efficiency and consistency.

Automatic elevator system for maintenance services

This work presents an electrically assisted automatic elevator system for maintenance of street lights. The system was installed in a public illumination maintenance truck. It consists of a telescopic column, a base structure, and a platform. The whole operation is monitored and controlled by a PLC. The system aims at reducing efforts in maintenance tasks as well as improving the work conditions for the electricians.

Adaptive automata for mapping unknown environments by mobile robots

Robotic mapping is one of the most important requirements for a truly autonomous mobile robot. Mobile robots should be able to building abstract representation of the physical environment, in order to navigate and work in such environment. This paper presents an adaptive way to make such representation. The proposed system allows the robot to explore all the environment and acquire the information incoming from the sensors (presence or absence of obstacles) while it travels. The robot may start the mapping process at any point of the space to be mapped. Due to the adaptability of the chosen method, the process has the capability of dynamically increase the memory requirements according to the already mapped area, even without any a priori knowledge of the environment.

Towards an IoT-based system for Smart City

Along with the breakthrough of the Internet of Things, a new city concept has attracted the attention of researchers, businesses and governments; it is the concept of Smart Cities. The purpose of this article is to present an IoT-based System for Smart Cities, called SIGINURB. The purpose of this System is to develop a range of applications that offers new services to students, employees, businesses and public administration in the University of São Paulo, raising their quality of life.

WTA 2007 - III.1 - Adaptive Automata Applied on Automation and Robotics (A4R)

This paper presents the researches conducted using Adaptive Automata (AA) theory in robotics equipments, process and systems. The generalization and behavior adaptation features of AA allow their application for process modeling, identification, classification and decision making in the same way as traditional computational logic such as Fuzzy Logic and Artificial Neural Networks. As sample of AA applications, the paper presents the research on pattern recognition and the research on hybrid mapping for robot navigation.

Communication Protocols for Application in Agricultural Vehicles

The use of electronic and computer systems for monitoring and control of agricultural machines, equipment, facilities and processes, and for the management of farms and agricultural companies has played a key role for the achievement of new and better standards in world agriculture. Besides increasing productivity, other objectives are decreasing the losses, rationally using the inputs, and protecting the environment. The advances in electronics have contributed to the great reduction in cost and price of equipment, at the same time that they have increased performance and reliability. The agricultural machines industry has been incorporating electronics to their products with the aim of providing more information to the operator about machine performance; registering that information for future analysis; automatically controlling the machines, freeing operator attention to other tasks, optimizing the use of the machine and avoiding unnecessary wear; and optimizing the use of inputs. Classical examples of those systems are on-board computers for tractors, which monitor variables of the engine, transmission and slip, in order to inform the operator or to control them automatically. Nowadays many machines and implements, such as combines (harvesters), planters and sprayers have monitoring and/or control systems which compensate the influence of the variation of true ground speed on the actual application rate of inputs (such as fertilizers or herbicides) or on the rate of harvesting. In order to do that, those machines need many different sensors and actuators, and a great amount of data must be processed and stored. Recently, with the new paradigm of precision agriculture, the use of embedded electronic systems in agricultural machines has significantly increased. Precision agriculture is a new management approach based on the fact that within the usual field management unit there can be a lot of variability in soil parameters, in the crop stand and in other agents such as pests and diseases, which can help explain (and eventually help compensate or correct) the yield variability within that field. The goals of this new paradigm are: to apply the inputs at variable rate within the field, each point receiving only what is effectively necessary (as opposed to a homogeneous application); to optimize the use of inputs, improving the cost/benefit ratio; to decrease the excess of inputs and the environmental impact of agriculture (Saraiva; Cugnasca, 1998). Precision agriculture is strongly based on automation, as it depends heavily on electronic data acquisition systems, on accurate positioning systems in the field (such as the Global Positioning System — GPS), and on variable rate application controllers. A great number of such systems and devices have been proposed for different field operations and parameters.

A Weighing System for Grab Loaders for Sugar Cane Yield Mapping

There are no commercial yield monitors for sugar cane despite the economic importance of the crop for sugar and alcohol production. As sugar cane is not mechanically harvested in some regions, special yield monitors need to be developed considering the harvest and load methods used. Grab loaders are essential when the crop is manually harvested. If the weight of the crop being loaded is monitored, yield maps can be produced. This paper presents the development of a weight measurement system for grab loaders. It is based on a mechanical device that contains a load cell adapted to the machine. As it is necessary to perform this measurement dynamically, the effect of movement on the measured value must be corrected. Thus, a mathematical model was developed to assess the effect of each parameter on the dynamics of the load and on the measured weight. In order to collect data to evaluate the model, an experimental measurement system was assembled and used in field tests. The analysis of the results showed that the model was able to describe the machine behaviour within the limits considered. Mean errors of ±2% or less were obtained for each load if the sampling period was adequately chosen. The frequency response of the angle sensors used and the low sampling rate of the subsystems used restricted the ability to define a single period for error minimisation. These two points will be addressed in the development of dedicated hardware, to allow the use of filtering and processing techniques in real time to obtain measurements with consistently low errors.

Double conical wheels based mobile robot for aerial power distribution lines inspection

There are several proposals of robots for inspection of overhead power network, and most of them make use of UAVs (Unmanned Aerial Vehicle). Such proposals present disadvantages like, low controllability, small payload capability and short operation range. On this scenario, this work presents a robot that can move over the power line and has as features: three driven wheels, distributed control system and self-configuration frame. The robot can overpass all power line devices taking the advantage of its wheels shape that provides a self-steering capability. The communications between each motor controller is done by IEEE 802.15.4 wireless communication standard. The robot can be controlled by an electrician to proceeds the inspection from the ground level or in an autonomous way. A prototype was constructed and was tested, in both, laboratory and field environment. The test results shown the effectiveness of the robot proposed.