Sebastian van Delden

A bluetooth-based architecture for android communication with an articulated robot

In industrial robotic environments, there are many different robots performing a variety of tasks. Each robot is controlled by its own teach pendant or via a networked socket application. However, to monitor the status or make minor changes to the programming of the robot, the user must obtain access to the teach pendant or terminal. In an effort to eliminate this need, this paper introduces an android platform that communicates with robots over a Bluetooth connection.

Retrieving NASA problem reports: a case study in natural language information retrieval

A system that retrieves problem reports from a NASA database is described. The database is queried with natural language questions. Part-of-speech tags are first assigned to each word in the question using a rule-based tagger. A partial parse of the question is then produced with independent sets of deterministic finite state automata. Using partial parse information, a look up strategy searches the database for problem reports relevant to the question. A bigram stemmer and irregular verb conjugates have been incorporated into the system to improve accuracy. The system is evaluated by a set of 55 questions posed by NASA engineers. A discussion of future research is also presented.

Pick‐and‐place application development using voice and visual commands

Purpose – The purpose of this paper is to design an interactive industrial robotic system which can be used to assist a “layperson” in re‐casting a generic pick‐and‐place application. A user can program a pick‐and‐place application simply by pointing to objects in the work area and speaking simple and intuitive natural language commands.Design/methodology/approach – The system was implemented in C# using the EMGU wrapper classes for OpenCV as well as the MS Speech Recognition API. The target language to be recognized was modelled using traditional augmented transition networks which were implemented as XML Grammars. The authors developed an original finger‐pointing algorithm using a unique combination of standard morphological and image processing techniques. Recognized voice commands trigger the vision component to capture what a user is pointing at. If the specified action requires robot movement, the required information is sent to the robot control component of the system, which then transmits the com...

Visual detection of objects in a robotic work area using hand gestures

A system for rapidly developing industrial robotic applications with visual hand gestures and voice commands is being developed. This paper reports on the hand gesture recognition component of the system which has been designed and implemented in C# using OpenCV. Seventy two test subjects interacted with the hand gesture component of the system and the empirical results of these experiments are reported. Screenshots of the hand gesture component are presented as well as known system limitations.

A FINITE STATE COMMA TAGGER

A method has been developed and implemented that assigns syntactic roles to commas. Text that has been tagged using a part-of-speech tagger serves as the input to the system. A set of Finite State Automata first assigns temporary syntactic roles to each comma in the sentence. A greedy learning algorithm is then used to determine the final syntactic roles of the commas. The system requires no training and is not domain specific. The performance of the system on numerous corpora is given and compared against a rule-based approach.

Finger tip tracking for manipulator jogging using the kinect

This paper reports on an interactive Human-Robot system which uses the Microsoft Kinect to track a human finger tip. By pointing to target locations in the robotic manipulators workarea, a user can intuitively and rapidly develop certain types of industrial robotic applications. The efficacy of the system is limited by the Kinects ability to precisely identify the finger tip in 3D space. Such Human-Computer Interactive devices are typically used to track large movements/gestures by a user in gaming and other applications. Empirical results are reported which show that the approach is able to precisely identify 3D location data and that the accuracy of the system is limited to the width of the users finger.

Problematic Syntactic Patterns

Several re-occurring problematic syntactic patterns which were encountered during theimplementation of a partial parser and natural language information retrieval system arepresented in this paper. These patterns cause syntax-based partial parsers, which rely oninitial part-of-speech tags, to make errors. We analyze two types of partial parsing errors:1) errors due to incorrect part-of-speech tags, and 2) errors made even though the parts-of-speech have been identified and present some novel solutions to avoiding these errors.

Constructing a Simple Visually-Guided Robotic Part-Grasping System with Off-the-Shelf Components

A novel approach to grasping and recognizing parts in an industrial robotic work cell is presented in this paper. The centroid, orientation, and length of elongated parts lying on a flat work area are estimated by a sequence of simple algorithms. Off-the-shelf components and freely downloadable software APIs make this system inexpensive and easily implemented. The approach has been implemented and tested with a Staubli RX60 manipulator. Results and future research are presented

Simple but useful algorithms for identifying noun phrase complements of embedded clauses in a partial parse

Two algorithms for identifying noun phrase complements of embedded clauses in a partial parse are presented. The candidate noun phrases play subject or object roles in (reduced) relative and infinitival clauses. The algorithms are tested on several sources and results are presented.

Improving Inter-level Communication in Cascaded Finite-State Partial Parsers

An improved inter-level communication strategy that enhances the capabilities of cascaded finite-state partial parsing systems is presented. Cascaded automata are allowed to make forward calls to other automata in the cascade as well as backward references to previously identified groupings. The approach is more powerful than a design in which the output of the current level is simply passed to the next level in the cascade. The approach is evaluated on randomly extracted sentences from the Encarta encyclopedia. A discussion of related research is also presented.