Knowledge assessment-practical example in testing

Knowledge assessment is inseparable part of current e-learning technologies. It can be used for self-assessment of students to give them feedback about their progress in a study or for an intermediate or final grading for tutors. However, knowledge tests are not developed with the adequate care. Author＇s experience in the area of knowledge assessment led to a confidence the ＂unstructured＂ testing is usually used in this process. It means that many of knowledge tests are not designed to reveal the reached level of knowledge. Moreover, testing suites are reviewed very seldom regarding their validity and items correlation. This paper presents experiences gained during the design and implementation of specific software focused on teaching several principles of the Unix-like operating systems. The structure of the specific assignment follows the Bloom＇s taxonomy of educational objectives.     

Optimizing Deep Learning Parameters Using Genetic Algorithm for Object Recognition and Robot Grasping

The performance of deep learning (DL) networks has been increased by elaborating the network structures. However, the DL netowrks have many parameters, which have a lot of influence on the performance of the network. We propose a genetic algorithm (GA) based deep belief neural network (DBNN) method for robot object recognition and grasping purpose. This method optimizes the parameters of the DBNN method, such as the number of hidden units, the number of epochs, and the learning rates, which would reduce the error rate and the network training time of object recognition. After recognizing objects, the robot performs the pick-and-place operations. We build a database of six objects for experimental purpose. Experimental results demonstrate that our method outperforms on the optimized robot object recognition and grasping tasks.     

Novel Biological Based Method for Robot Navigation and Localization

The capability and reliability are crucial characteristics of mobile robots while navigating in complex environments. These robots are expected to perform many useful tasks which can improve the quality of life greatly. Robot localization and decision-making are the most important cognitive processes during navigation. However, most of these algorithms are not efficient and are challenging tasks while robots navigate through complex environments. In this paper, we propose a biologically inspired method for robot decision-making, based on rat’s brain signals. Rodents accurately and rapidly navigate in complex spaces by localizing themselves in reference to the surrounding environmental landmarks. Firstly, we analyzed the rats’ strategies while navigating in the complex Y-maze, and recorded local field potentials (LFPs), simultaneously. The recorded LFPs were processed and different features were extracted which were used as the input in the artificial neural network (ANN) to predict the rat’s decision-making in each junction. The ANN performance was tested in a real robot and good performance is achieved. The implementation of our method on a real robot, demonstrates its abilities to imitate the rat’s decision-making and integrate the internal states with external sensors, in order to perform reliable navigation in complex maze.     

Humanoid robot arm performance optimization using multi objective evolutionary algorithm

As humanoid robots are expected to operate in human environments they are expected to perform a wide range of tasks. Therefore, the robot arm motion must be generated based on the specific task. In this paper we propose an optimal arm motion generation satisfying multiple criteria. In our method, we evolved neural controllers that generate the humanoid robot arm motion satisfying three different criteria; minimum time, minimum distance and minimum acceleration. The robot hand is required to move from the initial to the final goal position. In order to compare the performance, single objective GA is also considered as an optimization tool. Selected neural controllers from the Pareto solution are implemented and their performance is evaluated. Experimental investigation shows that the evolved neural controllers performed well in the real hardware of the mobile humanoid robot platform.     

Evolution of low-complexity neural controllers based on multiobjective evolution

In this paper, we present a new method based on multiobjective evolutionary algorithms to evolve low-complexity neural controllers for agents that have to perform multiple tasks simultaneously. In our method, each task and the structure of the neural controller are considered as separated objective functions. We compare the results of two different encoding schemes: (1) connectionist encoding, and (2) node-based encoding. The results show that multiobjective evolution can be successfully applied to generate low-complexity neural controllers. In addition, node-based encoding outperformed connectionist encoding in terms of agent performance and the robustness of the neural controller. This work was presented in part at the 12th International Symposium on Artificial Life and Robotics, Oita, Japan, January 25–27, 2007     

Multiobjective Evolution of Neural Controllers and Task Complexity

Robots operating in everyday life environments are often required to switch between different tasks. While learning and evolution have been effectively applied to single task performance, multiple task performance still lacks methods that have been demonstrated to be both reliable and efficient. This paper introduces a new method for multiple task performance based on multiobjective evolutionary algorithms, where each task is considered as a separate objective function. In order to verify the effectiveness, the proposed method is applied to evolve neural controllers for the Cyber Rodent (CR) robot that has to switch properly between two distinctly different tasks: 1) protecting another moving robot by following it closely and 2) collecting objects scattered in the environment. Furthermore, the tasks and neural complexity are analyzed by including the neural structure as a separate objective function. The simulation and experimental results using the CR robot show that the multiobjective-based evolutionary method can be applied effectively for generating neural networks that enable the robot to perform multiple tasks simultaneously.     

Guide robot intelligent navigation in urban environments

This paper presents an intelligent robotic system to guide visually impaired people in urban environments. The robot is equipped with two laser range finders, global positioning system (GPS), camera, and compass sensors. All the sensors data are processed by a single laptop computer. We have implemented different navigation algorithms enabling the robot to move autonomously in different urban environments. In pedestrian walkways, we utilize the distance to the edge (left, right, or both) to determine the robot steering command. In difference from pedestrian walkways, in open squares where there is no edge information, artificial neural networks map the GPS and compass sensor data to robot steering command guiding the visually impaired to the goal location. The neural controller is designed such as to be employed even in environments different from those in which they have been evolved. Another important advantage is that a single neural network controls the robot to reach multiple goal locations inside the open square. The proposed algorithms are verified experimentally in a navigation task inside the University of Toyama Campus, where the robot moves from the initial to goal location.     

Human-robot communication based on a mind model

We propose a mental image directed semantic theory (MIDST) and apply it to integrated multimedia information understanding, such as cross-media operations through intermediate knowledge representation. This article describes a multiagent model of the human mind based on the MIDST and its application to human–robot communication. This work was presented in part at the 10th International Symposium on Artificial Life and Robotics, Oita, Japan, February 4–6, 2005