We examine the implementation of clarification dialogues, a mechanism for ensuring that question answering systems take into account user goals by allowing them to ask series of related questions either by refining or expanding on previous questions with follow-up questions, in the context of open domain Question Answering systems. We develop an algorithm for clarification dialogue recognition through the analysis of collected data on clarification dialogues and examine the importance of clarification dialogue recognition for question answering. The algorithm is evaluated and shown to successfully recognize the start and continuation of clarification dialogues in 94% of cases. We then show the usefulness of the algorithm by demonstrating how the recognition of clarification dialogues can simplify the task of answer retrieval.

This paper deals with a method for robot navigation towards a moving goal. The goal maneuvers are not a priori known to the robot. Our method is based on the use of the kinematics equations of the robot and the goal combined with geometrical rules. First a kinematics model for the tracking problem is derived and two strategies are suggested for robot navigation, namely the velocity pursuit guidance law and the deviated pursuit guidance law. It turns out that in both cases, the robot's angular velocity is equal to the line of sight angle rate. Important properties of the navigation strategies are discussed and proven. In the presence of obstacles, two navigation modes are used: the tracking mode, which has a global aspect and the obstacle avoidance mode, which has a local aspect. In the obstacle avoidance mode, a polar diagram combining information about obstacles and directions corresponding to the pursuit is constructed. An extensive simulation study is carried out, where the efficiency of both strategies is illustrated for different scenarios.

This paper introduces conformal geometric algebra (CGA) for applications in computer vision and robotics. The authors show that CGA deals with our intuition and insight of the geometry and it helps us to reduce considerably the computational burden of the problems. The CGA can be applied not only to describe the geometry of the space, but to handle the algebra of incidence, as well as conformal transformations, to deal with kinematics or projective geometry problems. The authors show with real and simulated applications that this system can be of great advantage in robotics and computer vision.

In this paper, different adaptive control algorithms will be experimentally tested on a two axis SCARA type direct drive robot arm, and the performance of these algorithms will be compared. Being a direct drive system, the nonlinear effects, arising from the dynamics of the manipulator under high velocities, are directly reflected in the control of the manipulator. This makes the manipulator a more efficient test bed for testing the efficiency of the proposed adaptive schemes. In the experiments, we used fast trajectories rather than slow ones to observe how the proposed controllers compensate the dynamic nonlinear effects of manipulator dynamics. We will test some known adaptive control algorithms given in the literature along with our proposed adaptive control scheme which makes use of multiple models.

The paper developed a robust position tracking scheme with a neuro-fuzzy inverse model velocity controller of a tele-operated master-slave robot hand system for demining. And by using impedance force control and a two-posture recognition Neural Network (NN) a smooth grasping and releasing action is realized.

The paper presents a control scheme for simultaneous control of position and force of robot manipulator in contact with an elastodynamic environment. The control makes the assumption that interaction force between the robot and environment is adequately modeled by a second-order linear model with constant coefficients, and its implementation requires the knowledge of only boundary values of the environment parameters. It is shown that, provided that robot dynamics is exactly modeled, the scheme ensures asymptotic convergence of errors along nominal trajectories characterized by constant prescribed interaction forces and constant prescribed velocities along the contact surface.

This paper describes a project to implement necrophoric bee behaviour in a robot swarm. Pheromone communication is an effective means of coordinating the activities of insect colonies including food gathering, alarm and defense, reproduction and recognition of conspecifics. In a similar manner it is anticipated that pheromones will provide a valuable form of communication between robots. In order to investigate the problems and potential for this form of interaction, it was decided to implement an example of pheromone communication in a physical robotic system. Inspiration for this project came from the necrophoric behaviour of bees. The necrophoric pheromone released by dead bees triggers corpse removal behaviour in passing worker bees. In the context of a robot swarm one of the proposed applications for this behaviour is to locate and rescue disabled robots that release a pheromone as a form of distress signal. This paper provides details of the swarm robots used in the project, their sensors and the simple reactive control algorithm that was developed to mimic the necrophoric behaviour of bees. Results of practical experiments and simulations are also given.

The paper introduces artificial neural networks for the conventional control of robotic systems for better tracking performance. Different advanced dynamic control techniques are explained and a new second order recursive algorithm has been developed to tune the weights of the neural network. The problem of real-time control of a Pendubot system in difficult situations has been addressed. Examples, such as positioning and balancing structures, are presented and performances are compared to a conventional PD controller.

Suppose you're a corporate vice president at a well-known international software company, and you want to check on the visibility of one of your leading researchers in the outside world. You're sitting at your desk, so the most obvious thing to do is to enter their name into a search engine. If the well-known international software company happened to be Microsoft, and if the leading researcher happened to be Microsoft's Susan Dumais, and if the search engine you decided to use happened to be Google, you might be surprised to find that the sponsored link that comes atop the search results is actually from Google itself, exhorting you to ‘Work on NLP at Google’, and alerting you to the fact that ‘Google is hiring experts in statistical language processing’.

A task-oriented system structure has been developed. In normal industrial robot programming, the path is created and the process is based on the path. Here a process-focused method is proposed, where a task can be split in sub-tasks, one for each part of the process with similar process-characteristics. By carefully encapsulating the information needed to execute a sub-task, this component can be re-used whenever the actual sub-task occurs. Applications using system design do not change between simulation and actual shop floor runs and the system allows a mix of real- and simulated components during simulation and run-time.

This technical note deals with the way of organizing the work of a manipulator on a mobile robot. Every robotic mission is composed of several navigation and manipulation operations. In this research, the robot arm actions have been sequenced, once the mobile robot has stopped, with the help of the graphical specification tool Grafcet. The proposed method has been successfully implemented in the mobile manipulator RAM-2 for transporting videotapes.

This paper presents a design methodology to stabilize a class of multi-variant nonlinear system after a high disturbance occurs. It investigates application of Takagi-Sugeno type fuzzy controller (T-S-FC) to an inverted pendulum mechanism, actuated by an armature-controlled DC electrical motor.

Fuzzy controllers use heuristic information in developing design methodologies for control of non-linear dynamic systems. This approach eliminates the need for comprehensive knowledge and mathematical modeling of the system, and in cases of more complex systems, approximation and simplifications in order to achieve feasible mathematical model is not required.

The paper presents the stages of development of the Fuzzy Controller for an inverted pendulum by developing a two-input, Mamdani type system. It evaluates the performance of the system. Then a four-input T-S-FC type is developed. The research compares performances of each controller and presents the result of tests. A model for a DC motor is developed in this study, in order to measure the effect of time delays and response time caused by inherent properties of the physical system. The final part will demonstrate the complete operational system with the DC electrical motor included in the test system.

As part of a larger project to develop an aid for writers that would help to eliminate stylistic inconsistencies within a document, we experimented with neural networks to find the points in a text at which its stylistic character changes. Our best results, well above baseline, were achieved with time-delay networks that used features related to the author's syntactic preferences, whereas low-level and vocabulary-based features were not found to be useful. An alternative approach with character bigrams was not successful.

Using the notation of the wave variables, this paper introduces an autoregressive predictor, which forecasts the future values of the delay based on its previous values. Using this new knowledge, the teleoperation control system can be tuned to achieve a better and more practical performance. The validity of this modeling is first verified by actual experiments and then the results are used in simulated teleoperations.

The dynamic modeling and trajectory following issues are addressed for mobile modular manipulators. Simulations are performed to validate the proposed algorithms.

An analytical method for design and performance analysis of language models (LM) is described, and an example interactive software tool based on the technique is demonstrated. The LM performance analysis does not require on-line simulation or experimentation with the recognition system in which the LM is to employed. By exploiting parallels with signal detection theory, a profile of the LM as a function of the design parameters is given in a set of curves analogous to a receiver-operating-characteristic display.

A real-time joint trajectory generator for planar walking bipeds is proposed. This trajectory planner generates dynamically stable motion patterns by using a set of objective locomotion parameters as its input, and by tuning and exploiting the natural upper body dynamics. The latter can be determined and manipulated by using the angular momentum equation. Basically, trajectories for hip and swing foot motion are generated, which guarantee that the objective locomotion parameters attain certain prescribed values. Additionally, the hip trajectories are slightly modified such that the upper body motion is steered naturally, meaning that it requires practically no actuation. This has the advantage that the upper body actuation hardly influences the position of the Zero Moment Point. The effectiveness of the developed strategy is demonstrated by simulation results.

This paper presents a strategy for generating fault-tolerant gaits of hexapod walking robots. A multi-legged robot is considered to be fault-tolerant with respect to a given failure if it is capable of continuing its walking after the occurrence of a failure, maintaining its static stability. The failure concerned in this paper is a locked joint failure for which a joint in a leg cannot move and is locked in place. The kinematic condition for the existence of fault-tolerant gaits is derived for straight-line walking of a hexapod robot on even terrain. An algorithm for generating fault-tolerant gaits is described and, especially, periodic gaits are presented for forward walking of a hexapod robot with a locked joint failure. The leg sequence and the stride length formula are analytically driven based on gait study and robot kinematics. A case study on post-failure walking of a hexapod robot with the wave gait is shown to demonstrate the applicability of the proposed method.