URL: http://www.cs.bham.ac.uk/research/projects/cosy/deliverables/matrix/general-general/axs-fido.html
Last changed: 14 Dec 2005

General Information about Fido
Aaron Sloman

Use 'back' button to return to matrix
'The main goal of the CoSy project is to advance the science of cognitive systems through a multi-disciplinary investigation of requirements, design options and trade-offs for human-like, autonomous, integrated, physical (eg., robot) systems, including requirements for architectures, for forms of representation, for perceptual mechanisms, for learning, planning, reasoning and motivation, for action and communication.'
http://www.cognitivesystems.org/index.asp
The project proposal and workplan elaborated on this high level goal in Section 4, as follows:
The present proposal is directed at construction of embodied artifacts, such as robots, with advanced cognitive functions. The systems will be endowed with facilities for automatic interpretation of the environment (in terms of mapping of the environment, recognition of a large number of objects, etc.), adaptive acquisition of new skills and tasks in cooperation with a human user, methods for advanced manipulation to allow the system to perform extended missions on behalf of the users and reasoning methods to ensure advanced autonomy. The interpretation facilities are to be used both to ensure autonomy and to verbalise knowledge and mission parameters to a user. Advanced service robots can be used for a large range of tasks in our everyday environment ranging from vacuuming to clearing the table after dinner. The tasks can also be in terms of mobility support for elderly and handicapped. To be of true utility to an average citizen it is essential that the system has facilities for automatic adaptation to the environment of the user, it must also be able to adapt to the habits of the owner, it must be able to understand the importance and consequences of instructions, and it must be sufficiently flexible to learn new skills that it was not endowed with from the factory. An example scenario is outlined below.
  1. The Hartmut family decides to acquire a CoSy system to assist them in their everyday life. A system is acquired from the local Robs-R-Us chain of stores
  2. Upon arrival at home the system is unpackaged and booted up. Upon wakeup the CoSy system greets the owners and asks them to give their names (Helga and Oskar). The owners are asked to give it a name (Fido). The system then asks for a tour of the home. As they traverse the home it builds up a map of the house, and objects of particular interest are named to the system (this is the refrigerator),
  3. When asked to clean the table after dinner, Fido already has a built-in strategy for cleaning of tables, but it needs to know the location of the dishwasher (Oskar shows him the dishwasher - a Bosch 4590). Upon recognition of the dishwasher Fido downloads an operations manual from the Cosy corporate site. It then goes about its task.

    Oskar notices that Fido is removing cutlery and dishes but moves glasses to one side of the table. He asks: `Why are you leaving the glasses?' Fido answers: `My movements can be clumsy, so I'll put them in the dishwasher after all the objects that might break them.' ...

  4. ...

    (followed by various examples of interactions with adults children and various other entities in the house).

The purpose of this matrix

The main purpose of this matrix is to contribute to the analysis of requirements, leading to analysis of design options and trade-offs. As it is important to chose short term requirements and design goals that are relevant to the long term objectives we propose to do the analysis of requirements by having three sets of entries in the Matrix, namely
  1. Those requirements concerned with possible long term objectives for the Fido robot, constrained by the assumptions (a) that it will move around an ordinary house and (b) that it will in most ways not be more intelligent or knowledgeable than a typical 5 year old child (which in humans provides a basis for subsequently learning a great deal more).
  2. A simplified subset of the requirements which we believe might be met by the end of the project and which (a) require significant extensions to current understanding of the science of intelligent systems, (b) have a good chance of contributing to long term goals in (1).
  3. A further simplified subset of the requirements which we hope to be able to satisfy by month 30 and which will provide major stepping stones to the end of project goals.
Requirements of type (1) are flagged in the matrix with the label 'fido', those of type (2) with the label 'cosy', and those with type (3) with the label 'kitty'.

For example there are some general comments about perception and the relationships between perception and action in Fido here

Although we are now focusing on the PlayMate robot, it would be good to accommodate a similar analysis of requirements for Explorer in the same framework, so that we can identify commonalities and use them as a basis for later work on integrating PlayMate and Explorer.

For example, there are obvious overlaps between the ontology of spaces required between PlayMate and Explorer as sketched here.

We shall add items relevant to the Philosopher scenario when appropriate. Some of the decisions between things to leave out or include in 'cosy' or 'kitty' may depend on what we hope to do about the Philosopher capabilities.

Some requirements for CoSy.
A draft overview of requirements for Kitty. More detai for Kitty.


Use 'back' button to return to matrix