Skip to main content

Über dieses Buch

This book includes extended and revised versions of a set of selected papers from the Ninth International Conference on Informatics in Control Automation and Robotics (ICINCO 2012), held in Rome, Italy, from 28 to 31 July 2012. The conference was organized in four simultaneous tracks: Intelligent Control Systems and Optimization, Robotics and Automation, Systems Modeling, Signal Processing and Control and Industrial Engineering, Production and Management.

ICINCO 2012 received 360 paper submissions, from 58 countries in all continents. From these, after a blind review process, only 40 were accepted as full papers, of which 20 were selected for inclusion in this book, based on the classifications provided by the Program Committee. The selected papers reflect the interdisciplinary nature of the conference as well as the logic equilibrium between the four abovementioned tracks. The diversity of topics is an important feature of this conference, enabling an overall perception of several important scientific and technological trends.



Intelligent Control Systems and Optimization


Chapter 1. Adaptive Flux Observers and Rotor Speed Sensor Fault Detection in Induction Motors

The problem of detecting a rotor speed sensor fault in induction motor applications with load torque and rotor/stator resistances uncertainties is addressed. It is shown that in typical operating conditions involving constant rotor speed and flux modulus and non-zero load torque, a constant non-zero (sufficiently large) difference between the measured speed and the actual speed may be on-line identified by an adaptive flux observer which incorporates a convergent rotor resistance identifier and relies on the measured rotor speed and stator currents/voltages. Simulation and experimental results illustrate the effectiveness of the proposed solution and show satisfactory fault detection performances.
R. Marino, S. Scalzi, P. Tomei, C. M. Verrelli

Chapter 2. On Visual Analytics in Plant Monitoring

This chapter introduces methods from the field of visual analytics and machine learning which are able to handle high feature dimensions, timed systems and hybrid systems, i.e. systems comprising both discrete and continuous signals. Further, a three steps tool chain is introduced which guides the operator from the visualization of the normal behavior to the anomaly detection and also to the localization of faulty modules in production plants.
Tim Tack, Alexander Maier, Oliver Niggemann

Chapter 3. Global Optimization for 2D SLAM Problem

A globally optimal approach is proposed in this work for map-joining SLAM problem. Traditionally local optimization based approaches are adapted for SLAM problem but due to highly non-convex nature of the SLAM problem, they are susceptible to local minima. In this work, we have exploited the theoretical limit on the number of local minima. The proposed approach is not dependent upon the good initial guess whereas existing approaches in SLAM literature requires a good starting point for convergence to the basin of global minima. Simulation and real dataset results are provided to validate the robustness of the approach to converge to global minima. This chapter provides the robotics community to look into the SLAM problem with global optimization approach by guarantying the global optimal solution in a least square cost function particularly when covariance matrices are defined as spherical.
Usman Qayyum, Jonghyuk Kim

Chapter 4. Stochastic Models and Optimization Algorithms for Decision Support in Spacecraft Control Systems Preliminary Design

Technological and command-programming control contours of spacecraft are modelled with Markov chains. These models are used for the preliminary design of spacecraft control system effective structure with the use of special DSS. Corresponding optimization problems with algorithmically given functions of mixed variables are solved with a special stochastic algorithm called self-configuring genetic algorithm that requires no settings determination and parameter tuning. The high performance of the suggested algorithm is proved by the solving real problems of the control contours structure preliminary design.
Eugene Semenkin, Maria Semenkina

Chapter 5. A Heuristic Control Algorithm for Robust Internal Model Control with Arbitrary Reference Model

In this chapter the problem of Robust Internal Model Control is considered for the case of linear plants with nonlinear uncertain structure. The reference command is produced by an arbitrary reference model. A finite step Heuristic Algorithm is proposed in order to derive the controller parameters that guarantee robust performance under the proposed solvability conditions. The proposed controller is successfully applied to a hydraulic actuator uncertain model including uncertain parameters arising from changes of the operating conditions and other physical reasons. The satisfactory performance of hydraulic actuator variables for all the expected range of the actuator model uncertainties and external disturbances is illustrated via simulation experiments.
M. G. Skarpetis, F. N. Koumboulis, A. S. Ntellis

Chapter 6. A Multi-Signal Variant for the GPU-Based Parallelization of Growing Self-Organizing Networks

Among the many possible approaches for the parallelization of self-organizing networks, and in particular of growing self-organizing networks, perhaps the most common one is producing an optimized, parallel implementation of the standard sequential algorithms reported in the literature. In this chapter we explore an alternative approach, based on a new algorithm variant specifically designed to match the features of the large-scale, fine-grained parallelism of GPUs, in which multiple input signals are processed at once. Comparative tests have been performed, using both parallel and sequential implementations of the new algorithm variant, in particular for a growing self-organizing network that reconstructs surfaces from point clouds. The experimental results show that this approach allows harnessing in a more effective way the intrinsic parallelism that the self-organizing networks algorithms seem intuitively to suggest, obtaining better performances even with networks of smaller size.
Giacomo Parigi, Angelo Stramieri, Danilo Pau, Marco Piastra

Robotics and Automation


Chapter 7. Office Delivery Robot Controlled by Modular Behavior Selection Networks with Planning Capability

Recently, assistance service using mobile robots becomes one of important issues. Accordingly, studies on controlling the mobile robots are spreading all over the world. In this line of research, we propose a hybrid architecture based on hierarchical planning of modular behavior selection networks for generating autonomous behaviors of the office delivery robot. Behavior selection network is suitable for goal-oriented problems, but it is too difficult to design a monolithic behavior network to deal with complex robot control. We decompose it into several smaller behavior modules and construct sequences of the modules considering the sub-goals, the priority in each task and the user feedback. The feasibility of the proposed method is verified on both the Webot simulator and Khepera II robot in an office environment with delivery tasks. Experimental results confirm that a robot can achieve goals and generate module sequences successfully even in unpredictable and changeable situations, and the proposed planning method reduces the elapsed time during tasks by 17.5 %.
Young-Seol Lee, Sung-Bae Cho

Chapter 8. Worst-Case Performance Analysis in $$\ell _1$$ -norm for an Automated Heavy Vehicle Platoon

Based on model set identification and unfalsification, robust performance measured in peak-to-peak gain is analyzed for heterogeneous platoons, inter-vehicle communication delays and actuator uncertainties. The goal is to demonstrate that safe platooning with acceptable performance can be achieved by utilizing the services already available on every commercial heavy truck with automated gearbox. Experimental verification of a three vehicle platoon is also presented.
Gábor Rödönyi, Péter Gáspár, József Bokor, László Palkovics

Chapter 9. Visual SLAM Based on Single Omnidirectional Views

This chapter focuses on the problem of Simultaneous Localization and Mapping (SLAM) using visual information from the environment. We exploit the versatility of a single omnidirectional camera to carry out this task. Traditionally, visual SLAM approaches concentrate on the estimation of a set of visual 3D points of the environment, denoted as visual landmarks. As the number of visual landmarks increases the computation of the map becomes more complex. In this work we suggest a different representation of the environment which simplifies the computation of the map and provides a more compact representation. Particularly, the map is composed by a reduced set of omnidirectional images, denoted as views, acquired at certain poses of the environment. Each view consists of a position and orientation in the map and a set of 2D interest points extracted from the image reference frame. The information gathered by these views is stored to find corresponding points between the current view captured at the current robot pose and the views stored in the map. Once a set of corresponding points is found, a motion transformation can be computed to retrieve the position of both views. This fact allows us to estimate the current pose of the robot and build the map. Moreover, with the intention of performing a more reliable approach, we propose a new method to find correspondences since it is a troublesome issue in this framework. Its basis relies on the generation of a gaussian distribution to propagate the current error on the map to the the matching process. We present a series of experiments with real data to validate the ideas and the SLAM solution proposed in this work.
David Valiente, Arturo Gil, Lorenzo Fernández, Óscar Reinoso

Chapter 10. Metrics for Path Planning of Reconfigurable Robots in Uneven Terrain

In this chapter we present metrics for rough terrain motion planning used by our hierarchical planner. We employ a two-stage planning approach which allows us to use different cost functions for an initial path search and a detailed motion planning step. To quickly find an initial path we use a roughness quantification and the operating limits of the robot, which allow a fast assessment of the drivability. We then refine the initial path in rough regions of the environment by planning the complete robot states. To determine the desired robot configurations our newly developed metrics consider the system’s actuators, its safety and the time required for traversal. Real world experiments prove the validity and feasibility of the cost functions.
Michael Brunner, Bernd Brüggemann, Dirk Schulz

Chapter 11. A Combined Direct and Indirect Adaptive Control Scheme for a Wheeled Mobile Robot using Multiple Models

This chapter presents a method about trajectory tracking control of a nonholonomic wheeled mobile robot. The main focus of the chapter is to improve the transient response for the trajectory tracking control of mobile robots including dynamic parameter uncertainties. An adaptive combined direct and indirect control scheme is used for compensation of tracking errors in case of dynamic parameter uncertainties. The transient behavior for the adaptive tracking control is improved by a multiple models approach. The overall control system includes both a kinematic and dynamic controller. The kinematic controller produces linear and angular velocities required for mobile robot to track desired trajectory. The combined direct and indirect adaptive dynamic controller with adaptive multiple identification models takes these velocities as inputs and produces torques that will be applied to the robot. Simulation results indicate effectiveness of the proposed control scheme.
Altan Onat, Metin Ozkan

Chapter 12. Real-Time Visual Servoing Based on New Global Visual Features

This chapter proposes a new approach to achieve real-time visual servoing tasks. Our contribution consists in the definition of new global visual features as a random distribution of limited set of pixels luminance. The new method, based on a random process, reduces the computation time of the visual servoing scheme and removes matching and tracking process. Experimental results validate the proposed approach and show its robustness regarding to the image content.
Laroussi Hammouda, Khaled Kaaniche, Hassen Mekki, Mohamed Chtourou

Chapter 13. Compliance Error Compensation in Robotic-Based Milling

This chapter deals with the problem of compliance errors compensation in robotic-based milling. Contrary to previous works that assume that the forces/torques generated by the manufacturing process are constant, the interaction between the milling tool and the workpiece is modeled in details. It takes into account the tool geometry, the number of teeth, the feed rate, the spindle rotation speed and the properties of the material to be processed. Due to high level of the disturbing forces/torques, the developed compensation technique is based on the non-linear stiffness model that allows us to modify the target trajectory taking into account nonlinearities and to avoid the chattering effect. Illustrative example is presented that deals with robotic-based milling of aluminum alloy.
Alexandr Klimchik, Dmitry Bondarenko, Anatol Pashkevich, Sébastien Briot, Benoît Furet

Chapter 14. A Modified LGMD Based Neural Network for Automatic Collision Detection

Robotic collision detection is a complex task that requires both real time data acquisition and important features extraction from a captured image. In order to accomplish this task, the algorithms used need to be fast to process the captured data and perform real time decisions. Real-time collision detection in dynamic scenarios is a hard task if the algorithms used are based on conventional techniques of computer vision, since these arecomputationally complex and, consequently, time-consuming, specially if we consider small robotic devices with limited computational resources. On the other hand, neurorobotic models may provide a foundation for the development of more effective and autonomous robots, based on an improved understanding at the biological basis of adaptive behavior. Particularly, our approach must be inspired in simple neural systems, which only requires a small amount of neural hardware to perfom complex behaviours and, consequently, becomes easier to understand all the mechanism behind these behaviours. By this reason, flying insects are particularly attractive as sources of inspiration due to the complexity and efficiency of the behaviours allied with the simplicity of a reduced neural system. The Lobula Giant Movement Detector (LGMD) is a wide-field visual neuron located in the Locust optic lobe. It responds selectively to looming objects and can trigger avoidance reactions when a rapidly approaching object is detected. Based on the relatively simple encoding strategy of the LGMD neuron, different bio-inspired neural networks for collision avoidance were developed. In the work presented in this chapter, we propose a new LGMD model based on two previous models, in order to improve over them by incorporating other features. To accomplish this goal, we proceed as follows: (1) we critically analyse different LGMD models proposed in literature; (2) we highlight the convergence or divergence in the results obtained with each of the models; (3) we merge the advantages/disadvantages of each model into a new one. In order to assess the real-time properties of the proposed model, it was applied to a real robot. The obtained results have shown the high capability and robustness of the LGMD model to prevent collisions in complex visual scenarios.
Ana Carolina Silva, Jorge Silva, Cristina Peixoto dos Santos

Chapter 15. Vision Based Motion Estimation of Obstacles in Dynamic Unstructured Environments

Modeling static and dynamic traffic participants is an important requirement for driving assistance. Reliable speed estimation of obstacles is an essential goal especially when the surrounding environment is crowded and unstructured. In this chapter we propose a solution for real-time motion estimation of obstacles by using the pairwise alignment of object delimiters. Instead of involving the whole 3D point cloud, more compact polygonal models are extracted from a classified digital elevation map and are used as input data for the alignment process.
Andrei Vatavu, Sergiu Nedevschi

Chapter 16. Real-Time Vision-Based Pedestrian Detection in a Truck’s Blind Spot Zone Using a Warping Window Approach

In this chapter we present a vision-based pedestrian tracking system targeting a specific application: avoiding accidents in the blind spot zone of trucks. Existing blind spot safety systems do not offer a complete solution to this problem. Therefore we propose an active alarm system, which automatically detects vulnerable road users in blind spot camera images, and warns the truck driver about their presence. The demanding time constraint, the need for a high accuracy and the large distortion that a blind spot camera introduces makes this a challenging task. To achieve this we propose a warping window multi-pedestrian tracking algorithm. Our algorithm achieves real-time performance while maintaining high accuracy. To evaluate our algorithm we recorded several pedestrian datasets with a real blind spot camera mounted on a real truck, consisting of realistic simulated dangerous blind spot situations. Furthermore we recorded and performed preliminary experiments with datasets including bicyclists.
Kristof Van Beeck, Toon Goedemé, Tinne Tuytelaars

Chapter 17. A Proposal of Risk Indexes at Signalised Intersections for ADAS Aimed to Road Safety

Statistical data show that road intersections are one of the critical areas for accidents. The analyses reported are aimed at estimating risk indexes, which might be provided on-board, when vehicles approach a road intersection regulated by traffic lights, by an ADAS based on the use of the infrastructure-to-vehicle (I2V) or vehicle-to-infrastructure (V2I) communication systems. Two possible uses of the risk indexes can be identified: if data can be detected in real time, the driver could be informed on-board of a potentially hazardous situation, using algorithms to predict the dynamics of the vehicle on the basis of the data detected from the monitoring; the other use would be detecting—in case the vehicle were already within the dilemma zone—the lowest risk manoeuvre and providing the driver with a message on board. The chapter also reports the effects which might be generated by this ADAS application.
Bruno Dalla Chiara, Francesco Paolo Deflorio, Serena Cuzzola

Signal Processing, Sensors, Systems Modeling and Control


Chapter 18. A Component-Oriented Model for Wastewater Pumping Plants

A typical wastewater pumping plant comprises a screening process and a pumping process. The first process separates coarse material out of wastewater, while the second one boosts the wastewater toward a treatment facility. Appropriate component models for such plants are hardly found in literature. Indeed, there exist standard component models in all-purpose fluid simulation tool libraries; their generality, however, makes those models too complex to be used for wastewater pumping plants. The lack of models forces engineers to test their control scenarios on real implemented systems, which may lead to unexpected delays and painful costs. In this work, easily manageable component-oriented models are derived and applied to the modeling and simulation of a real wastewater pumping system. The model derived here is implemented in the component-oriented Modelica language, and it helps better understand the system dynamics. Thereby, a tool is provided for evaluating the performance of possible control schemes.
Mohamed Abdelati, Felix Felgner, Georg Frey

Chapter 19. A System Identification Framework for Modeling Complex Combustion Dynamics Using Support Vector Machines

Machine Learning is being widely applied to problems that are difficult to model using fundamental building blocks. However, the application of machine learning in powertrain modeling is not common because existing powertrain systems have been simple enough to model using simple physics. Also, black box models are yet to demonstrate sufficient robustness and stability features for widespread powertrain applications. However, with emergence of advanced technologies and complex systems in the automotive industry, obtaining a good physical model in a short time becomes a challenge and it becomes important to study alternatives. In this chapter, support vector machines (SVM) are used to obtain identification models for a gasoline homogeneous charge compression ignition (HCCI) engine. A machine learning framework is discussed that addresses several challenges for identification of the considered system that is nonlinear and whose region of stable operation is very narrow.
Vijay Manikandan Janakiraman, XuanLong Nguyen, Jeff Sterniak, Dennis Assanis


Weitere Informationen