Autonomous driving demo for chunk rules

This is a demo of a cognitive agent driving a car along a route in simulated road network for the UK town of Frome, Somerset, using data exported from Open Street Maps as XML, transformed into chunks, and indexed into a quad tree for efficient rendering. The aim of the demo is to show how a cognitive agent can dynamically manage a variety of cognitive tasks involving different kinds of reasoning. For more details, see chunks and rules. Other background below.

The map data is represented in terms of chunks for the map's bounding box, and multiple paths and points. Each road consists of one or more paths. Each path defines a sequence of points. Each point defines a location as latitude and longitude. Both paths and points may have additional metadata. Paths form a taxonomy reflecting the different categories of roads, e.g. trunk, primary, secondary, tertiary, unclassified, residential and so forth.

The spatial indexing uses quad chunks to form a tree that progressively divides the map's bounding box into four equal rectangles, which themselves are so divided, and so forth. Each quad has a bounding box, a set of subpaths, and a set of child quads. The root quad has the set of paths. The Cohen-Sutherland algorithm is used to split paths into subpaths that fit within a given quad's bounding box.


Dave Raggett <>

eu logo This work is supported by the European Union's Horizon 2020 research and innovation programme under grant agreement No 780732, project Boost 4.0