Search Results

New vehicle control algorithms are needed to meet future emissions and fuel economy mandates that are quite likely to require a measurement of ambient specific humidity (SH). Current practice is to obtain the SH by measurement of relative humidity (RH), temperature and barometric pressure with physical sensors, and then to estimate the SH using a fit equation. In this paper a novel approach is described: a system of neural networks trained to estimate the SH using data that already exists on the vehicle bus. The neural network system, which is referred to as a virtual SH sensor, incorporates information from the global navigation satellite system such as longitude, latitude, time and date, and from the vehicle climate control system such as temperature and barometric pressure, and outputs an estimate of SH. The conclusion of this preliminary study is that neural networks have the potential of being used as a virtual sensor for estimating ambient and intake manifold's SH.

Side swipe accidents occur primarily when drivers attempt an improper lane change, drift out of lane, or the vehicle loses lateral traction. Past studies of lane change detection have relied on vehicular data, such as steering angle, velocity, and acceleration. In this paper, we use three physiological signals from the driver to detect lane changes before the event actually occurs. These are the electrocardiogram (ECG), galvanic skin response (GSR), and respiration rate (RR) and were determined, in prior studies, to best reflect a driver's response to the driving environment. A novel system is proposed which uses a Granger causality test for feature selection and a neural network for classification. Test results showed that for 30 lane change events and 60 non lane change events in on-the-road driving, a true positive rate of 70% and a false positive rate of 10% was obtained.

Drivers often engage in secondary in-vehicle activity that is not related to vehicle control. This may be functional and/or to relieve monotony. Regardless, drivers believe they can safely do so when their perceived workload is low. In this paper, we describe a data acquisition system and machine learning based algorithms to determine perceived workload. Data collected were from on-road driving in light and heavy traffic, and individual physiological measures were recorded while the driver also performed in-vehicle tasks. Initial results show how the workload function can be personalized to an individual, and what implications this may have for vehicle design.