Martin Mellado

Application of a real time expert system platform for flexible autonomous transport in industrial production

Abstract In a manufacturing environment, material transport plays a key role for the production process efficiency. In Schoeller-Bregenz (SB), a textile manufacturing SME, induction wire guided vehicles performed typical material transports, without the flexibility to adapt transport routes to changes in the production process. To overcome these deficits in the RETRARO project 8 an Autonomous Guided Vehicle (AGV) was designed and tested in the production plant. To adapt to changing production configurations, RETRARO had to provide solutions for the following functionality regarding economical constraints: algorithms for efficient path planning and health monitoring; economic sensor systems to provide navigation and obstacle avoidance functions; Integration of free navigation capabilities with existing wire guidance systems. This paper presents a survey on the AGV system design and details on the implementation of the control system based on: the real time expert system REAKT for path planning and health monitoring; a low cost ultra sonic range detection system using sensor data fusion for navigation and obstacle avoidance.

Assessment of eggplant firmness with accelerometers on a pneumatic robot gripper

A gripper for handling and sorting eggplants by their firmness has been designed.An underactuation motion and a jamming system provide adaptation to eggplant shape.Inertial sensors attached to the gripper jaws collect decelerations during grasping.A software processes decelerations and provides 16 independent variables.A model has been developed and validated to explain firmness with the sensor data. A pneumatic robot gripper capable of sorting eggplants according to their firmness has been developed and tested. The gripper has three fingers and one suction cup. Each finger has an inertial sensor attached to it. One of the fingers adapts to and copies the shapes of eggplants when the jamming of its internal granular material changes from soft to hard. The other fingers adapt to the shape of the eggplant with the use of extra degrees of freedom. Specific software acquires and processes the information obtained with the inertial sensors and generates 16 independent variables extracted from the signals. A total of 234 eggplants were selected and tested on the same day with the robot gripper, during the pick-and-place operation, and with a destructive firmness tester. The non-destructive parameters extracted from the gripper finger accelerometers were used to build and validate a partial least square model, with a calibration regression coefficient of r=0.87 and a high prediction performance (r=0.90). Furthermore, from the results of the paper, it has been seen that the procedure can be simplified by using only two non-destructive impacts and one uniaxial accelerometer to assess eggplant firmness. The non-destructive assessment of firmness while grasping agricultural products in pick-and-place operations could be implemented in many prehensile pneumatic robot grippers. This technique could mean an important advance in the hygienic postharvest handling of fruits and vegetables.

Reactive planning simulation in dynamic environments with VirtualRobot

One of the difficulties of using Artificial Neural Networks (ANNs) to estimate atmospheric temperature is the large number of potential input variables available. In this study, four different feature extraction methods were used to reduce the input vector to train four networks to estimate temperature at different atmospheric levels. The four techniques used were: genetic algorithms (GA), coefficient of determination (CoD), mutual information (MI) and simple neural analysis (SNA). The results demonstrate that of the four methods used for this data set, mutual information and simple neural analysis can generate networks that have a smaller input parameter set, while still maintaining a high degree of accuracy.

Robot planning and re-planning using decomposition, abstraction, deduction, and prediction

Abstract One of the current topics of research in robotics is activity planning. Most works in this field are related to artificial intelligence (AI) techniques. The main feature of robot planning systems is that they are focused on the particular robot problem or are intended to solve a particular aspect of the planning problem. For this last case, motion planning or task planning are solved in an independent way. This paper covers robot activity planning and re-planning, from the high-level activity description to the final trajectory of the robot. Problem decomposition and abstraction are applied to obtain a robot plan; deduction and prediction are used for on-line trajectory re-planning. Temporal reasoning enables the obtention of velocity tuning for a trajectory.

A Robot-Soccer-Coordination Inspired Control Architecture Applied to Islanded Microgrids

Nowadays, islanded microgrids present a high interest due to the increasing penetration of renewable energy resources, especially in remote areas, or for improving the local energy reliability. A microgrid can operate in grid-connected or islanded mode, being necessary the use of energy storage systems (ESSs) under islanded operation, in order to ensure the generation/consumption power balance and smooth uncertainties in the dynamics of the renewable energy sources (RESs). Particularly, in islanded operation at least one of the distributed energy resources (DERs) should assume the regulation of the common bus. In a microgrid, every DER may be able to cooperate with the grid regulation in accordance to its particular operational conditions. In this sense, a centralized unit with a global perception of the load demand, the power provided by the RESs, and the storage capacity of the ESSs, may ensure proper and reliable operation of the microgrid. This paper proposes a structured architecture based on tactics, roles, and behaviors for a coordinated operation of islanded microgrids. The architecture is inspired on a robot soccer strategy with global perception and centralized control, which determines the changes among operation modes for the DERs in an islanded ac microgrid.

Architecting centralized coordination of soccer robots based on principle solution

Coordination strategy is a relevant topic in multi-robot systems, and robot soccer offers a suitable domain to conduct research in multi-robot coordination. Team strategy collects and uses environmental information to derive optimal team reactions, through cooperation among individual soccer robots. This paper presents a diagrammatic approach to architecting the coordination strategy of robot soccer teams by means of a principle solution. The proposed model focuses on robot soccer leagues that possess a central decision-making system, involving the dynamic selection of the roles and behaviors of the robot soccer players. The work sets out from the conceptual design phase, facilitating cross-domain development efforts, where different layers must be interconnected and coordinated to perform multiple tasks. The principle solution allows for intuitive design and the modeling of team strategies in a highly complex robot soccer environment with changing game conditions. Furthermore, such an approach enables systematic realization of collaborative behaviors among the teammates. Graphical abstract

Strategy planning for collaborative humanoid soccer robots based on principle solution

Collaborative humanoid soccer robots are currently under the lime light in the rapidly advancing research area of multi-robot systems. With new functionalities of software and hardware, they are becoming more versatile, robust and agile in response to the changes in the environment under dynamic conditions. This work focuses on a new approach for strategy planning of humanoid soccer robot teams as in the RoboCup Standard Platform League. The key element of the approach is a holistic system model of the principle solution encompassing various strategies of a soccer robot team. The benefits of the model-based approach are twofold—it enables intuitive behavioral specification of the humanoid soccer robots in line with the team strategies envisaged by the system developers, and it systematizes the realization of their collaborative behaviors based on the principle solution. The principle solution is modeled with the newly developed specification technique CONSENS® for the conceptual design of mechatronic and self-optimizing systems.

A Car-Seat Example of Automated Anthropomorphic Testing of Fabrics Using Force-Controlled Robot Motions

For the last years, automation is widely used to relieve humans from repetitive tasks, primarily and firstly within manufacturing. However, for products with less ideal (or hard to model) properties, and when forces depends on human interaction, auto mated testing has not been explored until now. This work presents the analysis of the (human-dependent) motions/forces based on a fully implemented test case for car-seat testing. For emulation of the corresponding mechanical wear, an experimental test bench was developed. A sensor mat with a pressure gauge net was used in the test bench to determine the relevant loads, and the corresponding movements performed by the humans when sitting in a car seat were acquired by means of a photogrammetry system. Finally, to automate the reproduction of such movements by means of a dummy held by a robot, several controllers have been developed to regulate the force applied by the dummy on the seat. Simplicity and force-control performance for the human replication was also investigated in this work, showing the benefit of freely programmable (open) force control. The developed system has many practical applications, as allowing the analysis of the wearing caused by these movements on the seat upholstery. Thus, force controlled testing of fabrics using robots is a viable option.

Architecture Definition for a Flexible Transport System in Industrial Production

Abstract Most of mobile robots in industrial environments are wire guided, following induction wires and artificial reference marks laid down on the factory floor. This kind of approach has the lack of flexibility to adapt the system for different transport requirements. This lack may be overcome using Autonomous Guided Vehicles (AGVs) which must be provided with an intelligent control system. Precisely, the purpose of the RETRARO project was to support intelligent control of an AGV in a specific production process. In this paper, the architecture of the system is explained, making special attention to the different module functions and their interfaces.

Information Fusion in Multi-agent System Based on Reliability Criterion

The paper addresses the problem of information fusion in Multi-Agent System. Since the knowledge of the process state is distributed between agents, the efficiency of the task performance depends on a proper information fusion technique applied to the agents. In this paper we study the case in which each agent has its own sensing device and is able to collect information with some certainty. Since the same information can be detected by multiple agents, the global certainty about the given fact derives from the fusion of information exchanged by interconnecting agents. The key issue in the method proposed, is an assumption that each agent is able to asses its own reliability during the task performance. The method is illustrated by the pick-up-and-collection task example. The effectiveness of the method proposed is evaluated using relevant simulation experiments.