## A Speech Interface for a Mobile Robot Controlled by GOLOG (2000)

Citations: | 1 - 0 self |

### BibTeX

@MISC{Dylla00aspeech,

author = {Frank Dylla and Gerhard Lakemeyer},

title = {A Speech Interface for a Mobile Robot Controlled by GOLOG},

year = {2000}

}

### OpenURL

### Abstract

With today's high-level plan languages like GOLOG or rpl it is possible for mobile robots to cope with complex problems. Unfortunately, instructing the robot what to do or interacting with it is still awkward. Usually, instructions are given by loading the appropriate program and interacting amounts to little more than pressing buttons positioned on the robot itself.

### Citations

4273 | A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition
- Rabiner
- 1989
(Show Context)
Citation Context ...tworks [2] or distance functions [16]. Especially, the statistical interpretation of distance functions has been found very useful [8, 13], in particular in connection with Hidden Markov Models (HMM) =-=[24]-=-. An overview is given in [1, 19, 20, 14]. 4 The basic principles of a statistical speech recognizer are shown in Figure 2. Figure 2. The Basic Structure of a statistical speech recognizer The idea is... |

2649 |
Introduction to Statistical Pattern Recognitionâ€ť, 2nd edition
- Fukunaga
- 1990
(Show Context)
Citation Context ...nt methods to build a speech recognizer, for example, using neural networks [2] or distance functions [16]. Especially, the statistical interpretation of distance functions has been found very useful =-=[8, 13]-=-, in particular in connection with Hidden Markov Models (HMM) [24]. An overview is given in [1, 19, 20, 14]. 4 The basic principles of a statistical speech recognizer are shown in Figure 2. Figure 2. ... |

500 | GOLOG: A Logic Programming Language for Dynamic Domains
- Levesque, Reiter, et al.
- 1996
(Show Context)
Citation Context ...bot to perform tasks from prespecified domains like mail or coffee delivery. Limited forms of interaction are also supported. 1 Motivation and Goals With today's high-level plan languages (like GOLOG =-=[12, 15]-=- or rpl [18]) it is possible for mobile robots to cope with complex problems. Unfortunately, instructing the robot what to do or interacting with it is still awkward. Usually, instructions are given b... |

188 | The interactive museum tour-guide robot
- Burgard, Cremers, et al.
- 1998
(Show Context)
Citation Context ...ms of interaction are also supported. Among other things, the robot is able to deal with tasks like serving coffee, delivering letters, or guiding people through a museum like his older brother RHINO =-=[4]-=-. To get an idea of the awkwardness in interacting with RHINO, let us look at the museum guide application in a little more detail. During a tour visitors had to advise their requests to RHINO by butt... |

182 |
Continuous speech recognition by statistical methods
- Jelinek
- 1976
(Show Context)
Citation Context ...nt methods to build a speech recognizer, for example, using neural networks [2] or distance functions [16]. Especially, the statistical interpretation of distance functions has been found very useful =-=[8, 13]-=-, in particular in connection with Hidden Markov Models (HMM) [24]. An overview is given in [1, 19, 20, 14]. 4 The basic principles of a statistical speech recognizer are shown in Figure 2. Figure 2. ... |

163 |
Maximum Mutual Information Estimation of Hidden Markov Models Parameters for Speech Recognition
- Souza, V, et al.
- 1986
(Show Context)
Citation Context ...ns [16]. Especially, the statistical interpretation of distance functions has been found very useful [8, 13], in particular in connection with Hidden Markov Models (HMM) [24]. An overview is given in =-=[1, 19, 20, 14]-=-. 4 The basic principles of a statistical speech recognizer are shown in Figure 2. Figure 2. The Basic Structure of a statistical speech recognizer The idea is, roughly, to first find an abstract repr... |

108 | The mobile robot Rhino
- Buhmann, Burgard, et al.
- 1995
(Show Context)
Citation Context ...robot is also equipped with two Pentium PC's and a radio Ethernet link. 5.2 The Software The base software for indoor navigation is BeeSoft, which was developed initially as part of the RHINO-project =-=[3]-=- and which is still Figure 3. The mobile robot CARL continuously maintained and extended. The software by now consists of more than 25 modules. Figure 5.2 shows some of the key modules. Figure 4. The ... |

85 | An Incremental Interpreter for High-Level Programs with Sensing
- Giacomo, Levesque
- 1998
(Show Context)
Citation Context ...ecausesGOLOG programs were processed offline, that is, before executing any of the primitive actions. We have since moved to using indiGOLOG, developed at the University of Toronto, which is based on =-=[9, 5]-=- and which offers action to be executed in parallel and, perhaps more importantly, allows for dealing with sensing through online interpretation. Note that the above example involves a simple form of ... |

78 | A reactive plan language
- McDermott
- 1991
(Show Context)
Citation Context ...asks from prespecified domains like mail or coffee delivery. Limited forms of interaction are also supported. 1 Motivation and Goals With today's high-level plan languages (like GOLOG [12, 15] or rpl =-=[18]-=-) it is possible for mobile robots to cope with complex problems. Unfortunately, instructing the robot what to do or interacting with it is still awkward. Usually, instructions are given by loading th... |

62 | Reasoning about concurrent execution, prioritized interrupts, and exogenous actions in the situation calculus
- Giacomo, Lesperance, et al.
- 1997
(Show Context)
Citation Context ...ecausesGOLOG programs were processed offline, that is, before executing any of the primitive actions. We have since moved to using indiGOLOG, developed at the University of Toronto, which is based on =-=[9, 5]-=- and which offers action to be executed in parallel and, perhaps more importantly, allows for dealing with sensing through online interpretation. Note that the above example involves a simple form of ... |

40 |
Fundamentals of Speech Recognition. Englewood Cliffs, NJ: Prentice-Hall
- Rabiner, Juang
- 1993
(Show Context)
Citation Context ...t we first apply Bayes' Rule and obtain 5 R(x T 1 ) = argmaxw 1 :::w N P (x T 1 jw N 1 )P (w N 1 ) 4 We remark that there also are non-statistical methods like Nearest Neighborhood or Neural Networks =-=[23, 26]-=- 5 As usual, the denominator of Bayes' Rule is ignored. Hence we need to determine two probability distributions: ffl the conditional probability distribution P (x T 1 jw N 1 ); this is called the aco... |

31 |
Data driven search organization for continuous speech recognition
- Noll, Ney, et al.
- 1992
(Show Context)
Citation Context ...ns [16]. Especially, the statistical interpretation of distance functions has been found very useful [8, 13], in particular in connection with Hidden Markov Models (HMM) [24]. An overview is given in =-=[1, 19, 20, 14]-=-. 4 The basic principles of a statistical speech recognizer are shown in Figure 2. Figure 2. The Basic Structure of a statistical speech recognizer The idea is, roughly, to first find an abstract repr... |

28 | Golex: Bridging the gap between logic (golog) and a real robot
- HHNEL, BURGARD, et al.
- 1998
(Show Context)
Citation Context ...bot to perform tasks from prespecified domains like mail or coffee delivery. Limited forms of interaction are also supported. 1 Motivation and Goals With today's high-level plan languages (like GOLOG =-=[12, 15]-=- or rpl [18]) it is possible for mobile robots to cope with complex problems. Unfortunately, instructing the robot what to do or interacting with it is still awkward. Usually, instructions are given b... |

25 |
An introduction to hybrid HMM/connectionist continuous speech recognition
- MORGAN, BOURLARD
- 1995
(Show Context)
Citation Context ...ns [16]. Especially, the statistical interpretation of distance functions has been found very useful [8, 13], in particular in connection with Hidden Markov Models (HMM) [24]. An overview is given in =-=[1, 19, 20, 14]-=-. 4 The basic principles of a statistical speech recognizer are shown in Figure 2. Figure 2. The Basic Structure of a statistical speech recognizer The idea is, roughly, to first find an abstract repr... |

23 |
An interprocess communication system for building robotic architectures. Programmerâ€™s guide to version 10.xx
- Fedor, TCX
- 1993
(Show Context)
Citation Context ...ules. Figure 5.2 shows some of the key modules. Figure 4. The structure of BeeSoft The modules run asynchronously without a central control unit. Communication is handled by a module called tcxServer =-=[7]-=-. The arrows indicate the direction of communication as handled by TCX. Certain dependencies are also observed. For example, the planning module (PLAN) is not able to initiate a robot move without the... |

23 | State based Gaussian selection In large vocabulary continuous speech recognition using HMMs
- Gales, Knill, et al.
- 1999
(Show Context)
Citation Context |

19 |
Acoustic Modeling of Phoneme Units for Continuous Speech Recognition
- Ney
- 1990
(Show Context)
Citation Context ...o the details here except to mention that Hidden Markov Models are employed for the acoustic model and the linguistic model is often simplified by using so-called uni-, bi- or trigram-models are used =-=[21]-=-. For example, the trigram-model is defined as: P (w N 1 ) = N Y n=1 P (wn jwn\Gamma2 ; wn\Gamma1 ) Typical problems when dealing with speech recognition are: ffl Velocitity of speech, which is heavil... |

4 |
Neural networks or hidden markov models for automatic speech recognition: Is there a choice
- Bridle
- 1992
(Show Context)
Citation Context ...few remarks on the state of the implementation and some conclusions. 2 Statistical Speech Recognition There are many different methods to build a speech recognizer, for example, using neural networks =-=[2]-=- or distance functions [16]. Especially, the statistical interpretation of distance functions has been found very useful [8, 13], in particular in connection with Hidden Markov Models (HMM) [24]. An o... |

4 |
Situations, actions and causal laws (Memo 2
- McCarthy
- 1963
(Show Context)
Citation Context ...sible, provided the user is aware of the limitations and sticks to simple instructions. 4 GOLOG GOLOG [15] is a logical language for programming complex actions. It is based on the situation calculus =-=[17, 22]-=-, a dialect of second-order logic which we briefly introduce first. We will not go over the language in detail except to note the following features: all terms in the language are one of three sorts, ... |

3 |
Pirri F., Reiter R., Foundations for the Situation Calculus
- Levesque
- 1998
(Show Context)
Citation Context ...sible, provided the user is aware of the limitations and sticks to simple instructions. 4 GOLOG GOLOG [15] is a logical language for programming complex actions. It is based on the situation calculus =-=[17, 22]-=-, a dialect of second-order logic which we briefly introduce first. We will not go over the language in detail except to note the following features: all terms in the language are one of three sorts, ... |

3 |
Generating word hypotheses in continuous speech
- Niemann
- 1986
(Show Context)
Citation Context ...t we first apply Bayes' Rule and obtain 5 R(x T 1 ) = argmaxw 1 :::w N P (x T 1 jw N 1 )P (w N 1 ) 4 We remark that there also are non-statistical methods like Nearest Neighborhood or Neural Networks =-=[23, 26]-=- 5 As usual, the denominator of Bayes' Rule is ignored. Hence we need to determine two probability distributions: ffl the conditional probability distribution P (x T 1 jw N 1 ); this is called the aco... |

2 |
Design of a robust Speech Control System for mobile Robots, Master thesis
- Dylla
- 2000
(Show Context)
Citation Context ...still awkward. Usually, instructions are given by loading the appropriate program and interacting amounts to little more than pressing buttons positioned on the robot itself. The goal of this project =-=[6]-=- is to offer a robust and easily expandable speech interface for GOLOG, implemented on a mobile robot. 2 Using a headset, a user can instruct the robot to perform tasks from prespecified domains. Limi... |

2 |
Implementation und Comparison of Discriminative Methods for Speechrecognition with small Vocabulary
- Macherey
- 1998
(Show Context)
Citation Context ...f the implementation and some conclusions. 2 Statistical Speech Recognition There are many different methods to build a speech recognizer, for example, using neural networks [2] or distance functions =-=[16]-=-. Especially, the statistical interpretation of distance functions has been found very useful [8, 13], in particular in connection with Hidden Markov Models (HMM) [24]. An overview is given in [1, 19,... |

1 |
GOLEX: Ein Laufzeitsystem fr die Aktionsbeschreibungssprache GOLOG zur Steuerung des mobilen Roboters RHINO, Diplomarbeit, Rheinische Friedrich-Wilhelms-Universitt
- Haehnel
- 1998
(Show Context)
Citation Context ...ho wants coffee. If no one wants one, CARL will wait in his room until the next request is made. The feasibilty to use GOLOG to control a robot in a realistic setting was demonstrated in principle in =-=[4, 10]-=-. There the originalsGOLOG [15] was used whose downside was that any form of sensing or interaction had to be handled outside the formalism becausesGOLOG programs were processed offline, that is, befo... |