Weighting the path continuation in route planning
Proceedings of the 9th ACM international symposium on Advances in geographic information systems
Pictorial and Verbal Tools for Conveying Routes
COSIT '99 Proceedings of the International Conference on Spatial Information Theory: Cognitive and Computational Foundations of Geographic Information Science
Elements of Good Route Directions in Familiar and Unfamiliar Environments
COSIT '99 Proceedings of the International Conference on Spatial Information Theory: Cognitive and Computational Foundations of Geographic Information Science
When and Why Are Visual Landmarks Used in Giving Directions?
COSIT 2001 Proceedings of the International Conference on Spatial Information Theory: Foundations of Geographic Information Science
Enriching Wayfinding Instructions with Local Landmarks
GIScience '02 Proceedings of the Second International Conference on Geographic Information Science
Wayfinding choremes-a language for modeling conceptual route knowledge
Journal of Visual Languages and Computing
A uniform handling of different landmark types in route directions
COSIT'07 Proceedings of the 8th international conference on Spatial information theory
Algorithms for reliable navigation and wayfinding
SC'06 Proceedings of the 2006 international conference on Spatial Cognition V: reasoning, action, interaction
Pictorial representations of routes: chunking route segments during comprehension
Spatial cognition III
Landmarks in OpenLS — a data structure for cognitive ergonomic route directions
GIScience'06 Proceedings of the 4th international conference on Geographic Information Science
A model for context-specific route directions
SC'04 Proceedings of the 4th international conference on Spatial Cognition: reasoning, Action, Interaction
Context-Aware Indoor Navigation
AmI '08 Proceedings of the European Conference on Ambient Intelligence
Easiest-to-reach neighbor search
Proceedings of the 18th SIGSPATIAL International Conference on Advances in Geographic Information Systems
Generating adaptive route instructions using hierarchical reinforcement learning
SC'10 Proceedings of the 7th international conference on Spatial cognition
Spatially-aware dialogue control using hierarchical reinforcement learning
ACM Transactions on Speech and Language Processing (TSLP)
Evaluating and minimizing ambiguities in qualitative route instructions
Proceedings of the 19th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems
On qualitative route descriptions: representation and computational complexity
IJCAI'11 Proceedings of the Twenty-Second international joint conference on Artificial Intelligence - Volume Volume Two
Giving advice to people in path selection problems
Proceedings of the 11th International Conference on Autonomous Agents and Multiagent Systems - Volume 1
A web-based evaluation framework for spatial instruction-giving systems
ACL '12 Proceedings of the ACL 2012 System Demonstrations
SIGDIAL '12 Proceedings of the 13th Annual Meeting of the Special Interest Group on Discourse and Dialogue
Hi-index | 0.00 |
Current applications for wayfinding and navigation assistance usually calculate the route to a destination based on the shortest or fastest path from the origin. However, numerous findings in cognitive science show that the ease of use and communication of route instructions depends on factors other than just the length of a route, such as the number and complexity of decision points. Building on previous work to improve the automatic generation of route instructions, this paper presents an algorithm for finding routes associated with the "simplest" instructions, taking into account fundamental principles of human direction giving, namely decision point complexity, references to landmarks, and spatial chunking. The algorithm presented can be computed in the same order of time complexity as Dijkstra's shortest path algorithm, O(n2). Empirical evaluation demonstrates that the algorithm's performance is comparable to previous work on "simplest paths," with an average increase of path length of about 10% compared to the shortest path. However, the instructions generated are on average 50% shorter than those for shortest or simplest paths. The conclusions argue that the compactness of the descriptions, in combination with the incorporation of the basic cognitive principles of chunking and landmarks, provides evidence that these instructions are easier to understand.