| 1. |
- Boyraz Baykas, Pinar, 1981, et al.
(författare)
-
Multi-sensor Driver Drowsiness Monitoring
- 2008
-
Ingår i: Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering. - 2041-2991 .- 0954-4070. ; 222:11, s. 2041-2062
-
Tidskriftsartikel (refereegranskat)abstract
- A system for driver drowsiness monitoring is proposed, using multi-sensor dataacquisition and investigating two decision-making algorithms, namely a fuzzy inference system(FIS) and an artificial neural network (ANN), to predict the drowsiness level of the driver.Drowsiness indicator signals are selected allowing non-intrusive measurements. The experi-mental set-up of a driver-drowsiness-monitoring system is designed on the basis of the sought-after indicator signals. These selected signals are the eye closure via pupil area measurement,gaze vector and head motion acquired by a monocular computer vision system, steering wheelangle, vehicle speed, and force applied to the steering wheel by the driver. It is believed that, byfusing these signals, driver drowsiness can be detected and drowsiness level can be predicted.For validation of this hypothesis, 30 subjects, in normal and sleep-deprived conditions, areinvolved in a standard highway simulation for 1.5 h, giving a data set of 30 pairs. For designing afeature space to be used in decision making, several metrics are derived using histograms andentropies of the signals. An FIS and an ANN are used for decision making on the drowsinesslevel. To construct the rule base of the FIS, two different methods are employed and comparedin terms of performance: first, linguistic rules from experimental studies in literature and,second, mathematically extracted rules by fuzzy subtractive clustering. The drowsiness levelsbelonging to each session are determined by the participants before and after the experiment,and videos of their faces are assessed to obtain the ground truth output for training thesystems. The FIS is able to predict correctly 98 per cent of determined drowsiness states(training set) and 89 per cent of previously unknown test set states, while the ANN has a correctclassification rate of 90 per cent for the test data. No significant difference is observed betweenthe FIS and the ANN; however, the FIS might be considered better since the rule base can beimproved on the basis of new observations.
|
|
| 2. |
- Ivarsson, Maria, 1977-, et al.
(författare)
-
Look-ahead control – consequences of a non-linear fuel map on truck fuel consumption
- 2009
-
Ingår i: Proceedings of the Institution of mechanical engineers. Part D, journal of automobile engineering. - London : Professional Engineering Publishing. - 0954-4070 .- 2041-2991. ; 223:10, s. 1223-1238
-
Tidskriftsartikel (refereegranskat)abstract
- Consequences of non-linearities in specific fuel consumption (SFC) of a heavy truck combustion engine are studied with focus on such small road gradients that a constant speed is optimal if the engine torque has an affine relation to fuelling. A quasi-static analysis gives valuable insights into the intrinsic properties of minimization of fuel consumption. Two objective functions are shown to give different optimal velocity trajectories on a constant road gradient, when the non-linearity in SFC is significant, a notation which is quantified. For a significant non-linearity, when a constraint is set to keep a final time, switching between twocharacteristic speeds is optimal. Alternatively, if consumed time, in addition to fuel consumption, is part of the objective function, then keeping to one constant speed is optimal also for significant non-linearities. However, the different optimal solutions still show similarities, since for a certain significant non-linearity a specific speed range determined by the characteristic velocities is shown to be unobtainable for both optimality criteria. Similarresults are obtained for a full dynamic model including a realistic fuel map and other realistic constraints.
|
|
| 3. |
- Nilsson, Ylva, et al.
(författare)
-
Weak knock characterization and detection for knock control
- 2009
-
Ingår i: Proceedings of the Institution of mechanical engineers. Part D, journal of automobile engineering. - 0954-4070 .- 2041-2991. ; 223:D1, s. 107-129
-
Tidskriftsartikel (refereegranskat)abstract
- Knock is a fundamental phenomenon in combustion engines, and knock control is central in any engine management system. Better understanding of knock, and weak knock in particular, is considered from two main aspects: knock detection and knock characterization. The aim of knock detectors is both to detect knock and to estimate the crank angle at knock onset. Focusing on weaker knock than before, it is shown that knock detectors and algorithms have to take into account other characteristics of knock traces than the standard model. It is shown that the best-performing knock detector of those investigated is one that supervises changes in signal variance, except for low signal-to-noise ratios where it is advantageous to use also the oscillation frequency. Regarding characterization, an important result is that in a wide range of intensities there is an almost linear dependence between the logarithmic normalized knock energy and the rate of cycles with knock. This means for example that a knock controller can use feedback on the rate of cycles with knock instead of knock intensity and vice versa; both can in combination with better detection provide possibilities for smoother and more anticipatory control schemes.
|
|
| 4. |
- Pohl, Jochen, 1971-, et al.
(författare)
-
A semi-automated parallel parking system for passenger cars
- 2006
-
Ingår i: Proceedings of the Institution of mechanical engineers. Part D, journal of automobile engineering. - 0954-4070 .- 2041-2991. ; 220:1, s. 53-65
-
Tidskriftsartikel (refereegranskat)abstract
- Car parking has been, and still is, a growing problem, with increasing vehicle sizes in the luxury segment as well as sport-utility vehicles. This is especially true when bearing in mind the confined parking spaces in parking lots and cities. While damage during parking generally does not cause any injury to the passengers, it is costly and annoying. Park assist systems are by no means new on the market, since passive systems which provide longitudinal guidance using ultrasonic distance sensors have been available on the market for a number of years.The system presented is a semi-automated approach to parallel parking problems, as they frequently occur in European and Asian cities. The challenge during the development of this system was to have as few components as possible added to a standard vehicle, seeking reuse of many of the already built-in functionalities. The result is a system that leaves the longitudinal control of the vehicle to the driver but automates the steering process, and even stops the vehicle when the final parking position is reached.
|
|
| 5. |
- Melander, Olle, et al.
(författare)
-
Nt-proANP in plasma, a marker of salt sensitivity, is reduced in type 2 diabetes patients.
- 2005
-
Ingår i: Journal of Internal Medicine. - : Wiley. - 1365-2796 .- 0954-6820. ; 257:3, s. 281-288
-
Tidskriftsartikel (refereegranskat)abstract
- Objective. We recently showed that plasma concentration of N-terminal atrial natriuretic peptide (Nt-proANP) is strongly directly related to salt sensitivity. The aims of the present study were to test (i) whether plasma concentration of N-terminal brain natriuretic peptide (Nt-proBNP) is related to salt sensitivity and (ii) whether Nt-proANP, as a marker of salt sensitivity, differs between type 2 diabetes patients and nondiabetic subjects without a history of coronary heart disease. Methods. Nt-proBNP was determined in 30 Swedish normal subjects with heredity for primary hypertension and salt sensitivity was defined as the difference between mean arterial blood pressure after 1 week on a high-salt diet (240 mmol day-1) and 1 week on a low-salt diet (10 mmol day-1). Nt-proANP was measured in 253 patients with type 2 diabetes and in 230 nondiabetic subjects aged 40–70 years, all without a history of coronary heart disease. Results. Amongst the 30 subjects, in whom salt sensitivity was directly measured, Nt-proBNP was not correlated with salt sensitivity (R = −0.18, P = 0.35). Nt-proANP (median, interquartile range) was lower in patients with type 2 diabetes (505, 387–661 pmol L-1) than in nondiabetic subjects (536, 421–696 pmol L-1) (P = 0.02). In a multiple regression analysis heart rate (P < 0.00001), diastolic blood pressure (P = 0.02) and diabetes status (P = 0.02) were inversely related whereas age (P < 0.00001), cystatin C (P = 0.0006), hypertension treatment (P = 0.002) and female sex (P = 0.006) were directly related to ln(Nt-proANP). Conclusion. In contrast to Nt-proANP, Nt-proBNP is not related to salt sensitivity. Salt sensitivity, as estimated by Nt-proANP, seems to be reduced in type 2 diabetes.
|
|
