RPN (Robotic Programming Network) is an initiative to bring existing remote robot laboratories to a new dimension, by adding the flexibility and power of writing ROS code in an Internet browser and running it in the remote robot with a single click. The code is executed in the robot server at full speed, i.e. without any communication delay, and the output of the process is returned back. Built upon Robot Web Tools, RPN works out-of-the-box in any ROS-based robot or simulator. This paper presents the core functionality of RPN in the context of a web-enabled ROS system, its possibilities for remote education and training, and some experimentation with simulators and real robots in which we have integrated the tool in a Moodle environment, creating some programming courses and make it open to researchers and students (http: //robotprogramming.uji.es).

The Robot Programming Network (RPN) is an initiative for creating a network of robotics education laboratories with remote programming capabilities. It consists of both online open course materials and online servers that are ready to execute and test the programs written by remote students. Online materials include introductory course modules on robot programming, mobile robotics and humanoids, aimed to learn from basic concepts in science, technology, engineering, and mathematics (STEM) to more advanced programming skills. The students have access to the online server hosts, where they submit and run their programming code on the fly. The hosts run a variety of robot simulation environments, and access to real robots can also be granted, upon successful achievement of the course modules. The learning materials provide step-by-step guidance for solving problems with increasing level of difficulty. Skill tests and challenges are given for checking the success, and online competitions are scheduled for additional motivation and fun. Use of standard robotics middleware (ROS) allows the system to be extended to a large number of robot platforms, and connected to other existing tele-laboratories for building a large social network for online teaching of robotics.

The need for intervention in underwater environments has increased in recent years but there is still a long way to go before AUVs (Autonomous Underwater Vehicles) will be able to cope with really challenging missions. Nowadays, the solution adopted is mainly based on remote operated vehicle (ROV) technology. These ROVs are controlled from support vessels by using unnecessarily complex human–robot interfaces (HRI). Therefore, it is necessary to reduce the complexity of these systems to make them easier to use and to reduce the stress on the operator. In this paper, and as part of the TWIN roBOTs for the cooperative underwater intervention missions (TWINBOT) project, we present an HRI (Human-Robot Interface) module which includes virtual reality (VR) technology. In fact, this contribution is an improvement on a preliminary study in this field also carried out, by our laboratory. Hence, having made a concerted effort to improve usability, the HRI system designed for robot control tasks presented in this paper is substantially easier to use. In summary, reliability and feasibility of this HRI module have been demonstrated thanks to the usability tests, which include a very complete pilot study, and guarantee much more friendly and intuitive properties in the final HRI-developed module presented here.

The latest research in neural networks demonstrates that the class imbalance problem is a critical factor in the classifiers performance when working with multi-class datasets. This occurs when the number of samples of some classes is much smaller compared to other classes. In this work, four different options to reduce the influence of the class imbalance problem in the neural networks are studied. These options consist of introducing several cost functions in the learning algorithm in order to improve the generalization ability of the networks and speed up the convergence process.

No abstract