Ana Cruz-Martín

Adaptable Web interfaces for networked robots

Most research in networked robots that use Web interfaces for robot control has been focused recently on the network part, since Ethernet involves poor (unpredictable) time performance. However, we believe that the problem to be addressed is more general and should not be restricted only to communication engineering: the interfaced system as a whole should adapt to get the most from the user, from the connection, and from the robot, even when no strict performance is possible. For that purpose, this paper introduces a new architecture for Web remote operation of robots that exhibits a high degree of flexibility in its adaptation to each particular user (through modular, configurable JAVA applets), to the system time-varying performance (through probability-guided, run-time adaptation of control loops), and to the robot software architecture (the standard CORBA is assumed as its middleware). Our approach constitutes an initial step for adapting comprehensively to all the mentioned issues, hence permitting to be employed in very different scenarios: realtime control, telecare, remote surveillance, etc.

Log-Logistic Modeling of Sensory Flow Delays in Networked Telerobots

This paper deals with modeling the delays in the transmission of sensory data from networked telerobots, which would allow us to predict future times of arrival and thus provide guarantees on the time requirements of these distributed systems. Considering sequences of delays as a uni-dimensional signal, they easily exhibit rich stochastic behaviours - abrupt changes of regime and bursts -, due to the heterogeneity of the hardware and software components in the data path. There exist many approaches for modeling this kind of signals without explicit knowledge of the components: statespace methods, hidden Markov models, neural networks, etc., but they are mostly used for the stochasticity in the network components only. Besides, in the field of remote control, some knowledge about the controlled plant is assumed. Previously, we have proposed simpler statistical methods that do not require such complexity in the models or any plant knowledge, making them suitable for light weight implementations (e.g., in mobile phone interfaces). In this sense we reported elsewhere a lognormal three-parametrical model that fits well these delays as long as change detection is carried out appropriately. In this paper we propose a more flexible model: the log-logistic distribution, that has been found to fit delays better, although with higher computational cost. We also present an algorithm for modeling an entire delay signal based on it. Our results show quite good fittings of real datasets gathered from a number of real sensors, networks and application software.

Marginal Probabilistic Modeling of the Delays in the Sensory Data Transmission of Networked Telerobots

Networked telerobots are remotely controlled through general purpose networks and components, which are highly heterogeneous and exhibit stochastic response times; however their correct teleoperation requires a timely flow of information from sensors to remote stations. In order to guarantee these time requirements, a good on-line probabilistic estimation of the sensory transmission delays is needed. In many modern applications this estimation must be computationally highly efficient, e.g., when the system includes a web-based client interface. This paper studies marginal probability distributions that, under mild assumptions, can be a good approximation of the real distribution of the delays without using knowledge of their dynamics, are efficient to compute, and need minor modifications on the networked robot. Since sequences of delays exhibit strong non-linearities in these networked applications, to satisfy the iid hypothesis required by the marginal approach we apply a change detection method. The results reported here indicate that some parametrical models explain well many more real scenarios when using this change detection method, while some non-parametrical distributions have a very good rate of successful modeling in the case that non-linearity detection is not possible and that we split the total delay into its three basic terms: server, network and client times.

Log-logistic modeling of sensory flow delays in networked telerobots

This paper deals with the modeling of the delays in the transmission of sensory data coming from a networked telerobot, which would allow us to predict future times of arrival and provide guarantees on the time requirements of these systems. Considering these delay sequences as an uni-dimensional temporal signal, they easily exhibit rich stochastic behavior—abrupt changes of regime and bursts—due to the heterogeneity of the hardware and software components in the data path. There exist approaches for modeling this kind of signals without explicit knowledge of the system components: state-space reconstruction, hidden Markov models, neural networks, etc., but they are mostly focused on the stochasticity of the network only, without taking into account other elements in the sensory flow that also have an important influence in the delays. Previously, we have proposed simpler statistical methods that do not require any component knowledge either and are suitable for more lightweight implementations (e.g., in mobile phone interfaces). In this sense, we report elsewhere a log-normal three-parametrical model that fits reasonably well these delays as long as change detection is completely solved. Now we propose a more flexible solution: the log-logistic distribution, which has been found to fit delays better than the log-normal. In addition, we present two algorithms to model an entire delay signal, including abrupt nonlinearities, based on the log-logistic assumption. Our results show quite good fittings of real datasets gathered from a number of combinations of sensors, networks, and application software, provided that some mild assumptions hold.

A heterogeneity-enabled development system for educational mechatronics

Mechatronics is emerging as a leading trend in the current engineering scene. Hence, educational platforms are needed in order to train engineers for this new job market demand. At present time, the existing educational systems that deal with mechatronic aspects, like robotic, control or automation systems, lack of a systematic methodology of development. This deficiency translates into closed training platforms, where it is difficult to plug in new components at convenience. Extending any system with different high- and low-level components that allow the student to understand mechatronics from different perspectives is complicated within such approaches. This paper presents a development system that provides a basic model, framework and automated tools for building mechatronic platforms. Our system, that can be applied to different fields (like mobile and industrial robotics, automation systems, real-time systems, etc.), is also scalable, producing from simple mechatronics systems to complex ones. We describe a case of use where our development system has been used for educational robotics, combining LEGO Mindstorms with research-level robotic platforms.

Towards a Common Implementation of Reinforcement Learning for Multiple Robotic Tasks

Abstract Mobile robots are increasingly being employed for performing complex tasks in dynamic environments. Those tasks can be either explicitly programmed by an engineer or learned by means of some automatic learning method, which improves the adaptability of the robot and reduces the effort of setting it up. In this sense, reinforcement learning (RL) methods are recognized as a promising tool for a machine to learn autonomously how to do tasks that are specified in a relatively simple manner. However, the dependency between these methods and the particular task to learn is a well-known problem that has strongly restricted practical implementations in robotics so far. Breaking this barrier would have a significant impact on these and other intelligent systems; in particular, having a core method that requires little tuning effort for being applicable to diverse tasks would boost their autonomy in learning and self-adaptation capabilities. In this paper we present such a practical core implementation of RL, which enables the learning process for multiple robotic tasks with minimal per-task tuning or none. Based on value iteration methods, we introduce a novel approach for action selection, called Q-biased softmax regression (QBIASSR), that takes advantage of the structure of the state space by attending the physical variables involved (e.g., distances to obstacles, robot pose, etc.), thus experienced sets of states accelerate the decision-making process of unexplored or rarely-explored states. Intensive experiments with both real and simulated robots, carried out with the software framework also introduced here, show that our implementation is able to learn different robotic tasks without tuning the learning method. They also suggest that the combination of true online SARSA(λ) (TOSL) with QBIASSR can outperform the existing RL core algorithms in low-dimensional robotic tasks. All of these are promising results towards the possibility of learning much more complex tasks autonomously by a robotic agent.

Hierarchical regulation of sensor data transmission for networked telerobots

Networked telerobots carry sensors that send data, with stochastic transmission times, to a remote human operator, who must execute some real-time control task (e.g., navigation). In this paper we propose to regulate the sensory information being transmitted in order to guarantee soft real-time requirements and also optimize the quality of control, through a novel two-level hierarchical controller that both varies the amount of transmitted sensor data and dynamically reconfigures active sensors. Our controller has been implemented in a web-based teleoperator interface that is highly portable (it runs on desktop PCs, tablets, smartphones, etc.) and non-invasive, i.e., requires minimal modifications in the existing components of the system, thus being suitable for many applications. Here we present our regulation methods and the results of some experiments. They demonstrate the maximization of the transmitted data while guaranteeing the real-time requirements.

Automatic Regulation of the Information Flow in the Control Loops of a Web Teleoperated Robot

The use of the World Wide Web for robot teleoperation is growing in the last years due mainly to the pervasiveness of Internet and Web browsers, although Web interfaces usually use Ethernet networks that exhibit time unpredictability. Most recent research in the area has been focused on improving time predictability of the network under delays, jitter, and no guaranteed bandwidth. However, we believe that: i) not only the network, but every component in the interfaced system exhibit time unpredictability; and ii) improving time predictability is not the only solution: adapting the interfaced system to unpredictable conditions is also a possibility. In this paper we consider a Web interfaced robot as a set of control loops and describe and implement a hysteresis controller for regulating the flow of information through the loops as a method to satisfy the system time requirements under some unpredictable and varying conditions. For demonstrating the goodness of our algorithm, we: a) compare it with a near-optimal one automatically generated through reinforcement learning, and b) show an implementation of the algorithm for the direct teleoperation of a service mobile robot, obtaining a better behavior than the same system without flow regulation.

A software framework for coping with heterogeneity in the shopfloor

Purpose – The CIM framework pursues the integration of components in a manufacturing enterprise by means of computer systems. This, however, may be obstructed due to heterogeneity in the field: programmable controllers, robots, sensors and actuators, etc. in communications: different kinds of networks and/or field buses; and in the programming tools for all these devices. Thus a solution is needed to integrate heterogeneous software/hardware components in a well‐defined and flexible fashion. This paper seeks to address these issues.Design/methodology/approach – This paper proposes a metalanguage, called H, and a set of tools that serve for designing, implementing, deploying, and debugging distributed heterogeneous software on the shopfloor. The metalanguange includes fault‐tolerance and real‐time mechanisms, among other features.Findings – The use of a framework that can integrate different software and hardware components enables the engineer to take advantage of the best features of each existing techno...

Genetic algorithms based multirobot trajectory planning

This paper presents a methodology for planning trajectories in a multirobot system. The planner takes into account the speed constraints acting on the vehicles, so it provides a safe speed profile for every robot in the system. As the planner is decoupled and prioritized, it is necessary to set up some priority assignment that defines the order in which trajectories are computed. In order to obtain near-optimal solutions for that performance, a multirobot trajectories scheduler, based on genetic algorithms, has been designed, developed and tested. The multirobot trajectory planning is being incorporated into a three-layer multirobot architecture, so the lack of flexibility related to the deliberative planning can be compensated