Signal Modelling and Hidden Markov Models for Driving Manoeuvre Recognition and Driver Fault Diagnosis in an urban road scenario
Paper in proceedings, 2007
Hidden Markov models (HMM) are used to identify a vehicle's manoeuvre sequence and its appropriateness for a given urban road driving situation. One of the novel aspects of this work has been the development of an efficient signal modelling approach to form a context-aware, flexible system which proved to respond well in urban road scenarios, especially in situations where the driver is likely to have an accident due to impaired performance. Another contribution has been to clarify how HMMs can be used not just to recognize vehicle manoeuvres but also to distinguish an impaired driver from a normal one in complex driving contexts. The system has worked well on simulator data and is about to be implemented in the real conditions of an urban trajectory.
Data analysis
System testing
Stochastic processes
Artificial neural networks
Hidden Markov models
Fault diagnosis
Signal analysis
Safety
System analysis and design
Road transportation