Chapter 3: Early Studies

Chapter overview and links to contents

Ship collision avoidance was a plausible area of study that interested one of the sponsors of the research. At the time of the study, other research was being carried out towards automatic collision avoidance, or collision avoidance support.

3.1 Maritime collision avoidance
3.1.1 The nature of collision avoidance
3.1.2 Representations in collision avoidance
3.1.2.1 The literature on possible revisions to the collision regulations
3.1.2.2 The evidence of differences between individual watchkeeping officers
3.1.2.3 Work on collision avoidance advice systems
The Liverpool system
The Plymouth system
3.1.2.4 Undocumented considerations
3.1.3 Difficulty in collection of data
3.1.4 Difficulty in simulation

The real problem was that collision avoidance did not stand up as a separate problem, outside the realistic situation of navigation (live or in simulators). This prompted the study of something completely different, to try to get away from these problems. This was based around machine pole-balancing, as this was of research interest at the time in the Turing Institute, since it was a simple example of a class of dynamically unstable systems.

The problems with this are covered in the final section.

3.2 Dynamic control and machine learning
3.2.1 Fundamental ideas in dynamic control
3.2.2 The BOXES approach to pole-balancing
3.2.2.1 Chambers & Michie's ideas for cooperation
3.2.2.2 Recent work with the pole-and-cart system
3.2.3 Representation in machine learning generally
3.2.4 Commentary on relevance