Driver stress is a growing problem in the today’s world which causes a deterioration of cognitive skills, resulting in poor driving and an increase in the likelihood of traffic accidents. Prediction models allow us to avoid or at least minimize the negative consequences of stress. The physiological signals of low frequency and voltages, after amplification and filtration can be easily transmitted with low noise interference in the signal. The prediction of the upcoming stress level is made by taking into account heartbeat, Eeg signals. Design of a bio-signal based wearable driver assist system requires a functional algorithm capable of mapping features extracted from time series data of physiological signals to a known mental / affective state. The paper presents a comprehensive analysis of extraction and signal processing techniques of features from physiological signals.
This paper proposes a wearable device-based system to monitor the abnormal conditions of a driver where the system measures the motional and physiological information of the driver using the developed wearable device on the wrist. Preprocessing is used to distinguish the valid signal parts of the measured signals, because various noises can occur in wearable sensors. .Selye was the first researcher to refer to the term stress in a biological context. According to Selye, stress includes an inappropriate physiological response to any kind of demand. Stress refers to the condition that occurs in response to certain stimuli (stressors).
The data presented categorize the major risk factors responsible for traffic accidents as follows (according to their impact):
• human error 92%
• cognitive errors 40.6%
• judgment errors 34.1%
• execution errors 10.3%
• other 15%
• vehicle malfunction 2.6%
• road/environmental conditions 2.6%
• other 2.8%
In the research literature, there are many proposed methods to measure and quantify a driver’s cognitive load and stress levels. The two major types of methods proposed to measure stress are questionnaires and physiological signals.
Questionnaires allow us to assess a large part of the population. However, the result is based on the subjective perception of the participant. One of the most important works of research is the perceived stress questionnaire. This measurement employs the subjective perception of things and the emotional reaction. This questionnaire can be used regardless of age, gender, or profession of the participants. Other questionnaires relevant to measuring stress include the stress appraisal measure, the impact of event scale, and the perceived stress scale .
The methods based on physiological signals allow us to objectively and continuously determine the stress level of a user. However, they require the use of sensors, increasing the cost and reducing the number of possible participants. In addition, these solutions can cause discomfort if they are intrusive or heavy. However, these problems are being minimized with the proliferation of wearable devices in recent years.
• The current methods of stress monitoring is done only in hospitals. There are no ways of measuring autonomously. The corresponding person has to be frequently examined for any symptoms. Then after examining, a measure is taken such as medication or any other physical activities. And sleep detectors are not so modernised.
• Not suitable for instant actions
• Need of going to hospitals or clinics frequently
Not suitable for all drivers who works for long hours of driving.
This paper deals with the idea of driver stress and sleep monitoring. Here the sensors and microcontroller is used for the application. The microcontroller atmega 8 is used. The microcontroller is the center of command and action. The sensors such as the eeg sensor, the heartbeat sensor, alcohol sensor and the infrared sensor are used. The outputs of these sensors are fed to the microcontroller’s adc. The microcontroller is programmed in such a way that if there is any abnormality in the received input, it has to take action such as switching the relay, sending an sms alert etc…. the parameters such as the heartbeat rate of the driver, eeg signal are taken into account for stress and sleep monitoring.
• Here the eeg sensor monitors the brain waves of the driver, and based on the heartbeat and eeg signal inputs the state of the driver is identified.
• Incase of any abnormalities, an sms alert is sent via the gsm module.
• The infrared sensor is used for obstacle detection, incase of any, the sensor will intimate the microcontroller atmega8 which inturn will stop the engine.
• The alcohol sensor and the heartbeat sensors also act in the same way.
• The sensors are connected to the analog to digital converters of the microcontroller where the analog signals are then converted to digital signals.
• The microcontroller Atmega8 acts as the central control unit. The gsm module is connected to the microcontroller which receives data from the controller. The microcontroller is programmed in such a way that upon data receiving from the sensors, an action has to be taken at the lcd, motor and the gsm.
• The system compose of 230 AC supply from the wall socket. Here a 12-0-12 stepdown transformer is used which converts the 230 v ac o 12 v ac. Followed by a rectifier circuit made of two diodes IN4007. Then a capacitor is added parallely for filtering purpose.
• A voltage regulator 7805 is used which converts/regulates the 12 v dc to 5 v. and this is used by the microcontroller.
• The lcd pins are connected to the C3,C4,C5 to the rs, rw, enable pins respectively and the data pins TO d2,d3,d4,d5.
• The relay is connected to the pb7 pin of the microcontroller.
• The motor is connected to the normally open pin of the relay.
The gsm is connected to the microcontroller via the max232(transmitter and receiver pins)
• ATmega8 microcontroller
Codevision AVR for Embedded C programming.
• Gain: 40000
• Range: ±41.25?V (with VCC = 3.3V)
• Bandwidth: 0.8-49Hz
• Consumption: ~3Ma Input Impedance:100GOhm
• CMRR: 100dB
• Single-channel sensor
• Bipolar differential measurement
• Pre-conditioned analog output
• Small form factor
• Raw data output
• Human-Computer Interaction
• Evoked potentials analysis
• Neuro feedback
• Sleep studies
• Neurophysiology studies
• Biomedical devices prototyping
The electroencephalography (EEG) sensor has been especially designed for both classic and localized EEG measurement. When a cap is too intrusive, only a limited number of channels are needed, or you’d like to synchronously record EEG and non-EEG biosignals, this is the perfect solution. The bipolar configuration, with two measurement electrodes detects the electrical potentials in the specific scalp region with respect to a reference electrode, which should be placed in a region of low muscular activity. The resulting signal is the amplified difference between these two signals, eliminating the common unwanted signals detected by the surfaces. Its convenient form factor enables a discrete placement in regions such as the forehead, occipital, and others.
The number of bits for each channel depends on the resolution of the Analog-to-Digital Converter (ADC); in BITalino the first four channels are sampled using 10-bit resolution, while the last two are sampled using 6-bit .
An alcohol sensor detects the attentiveness of alcohol gas in the air and an analog voltage is an output reading. The sensor can activate at temperatures ranging from -10 to 50° C with a power supply is less than 150 Ma to 5V. The sensing range is from 0.04 mg/L to 4 mg/L, which is suitable for breathalyzers.
MQ-135 Gas Sensor
The MQ-135 gas sensor senses the gases like ammonia nitrogen, oxygen, alcohols, aromatic compounds, sulfide and smoke. The boost converter of the chip MQ-3 gas sensor is PT1301. The operating voltage of this gas sensor is from 2.5V to 5.0V. The MQ-3 gas sensor has a lower conductivity to clean the air as a gas sensing material. In the atmosphere we can find polluting gases, but the conductivity of gas sensor increases as the concentration of polluting gas increases. MQ-135 gas sensor can be implementation to detect the smoke, benzene, steam and other harmful gases. It has potential to detect different harmful gases. The MQ-135 gas sensor is low cost to purchase. The basic image of the MQ-135 sensor is shown in the below figure.
MQ-135 Gas Sensor
Basic Pin Configuration of Alcohol Sensor:
The MQ-3 alcohol gas sensor consists of total 6-pins including A, H, B and the other three pins are A, H, B out of the total 6-pins we use only 4 pins. The two pins A, H are used for the heating purpose and the other two pins are used for the ground and power. There is a heating system inside the sensor, which is made up of aluminium oxide, tin dioxide. It has heat coils to produce heat, and thus it is used as a heat sensor. The below diagram shows the pin diagram and the configuration of the MQ-3 alcohol sensor.
Pin Configuration Of Alcohol Sensor
The MQ-135 alcohol sensor consists of a tin dioxide (SnO2), a perspective layer inside aluminium oxide micro tubes (measuring electrodes) and a heating element inside a tubular casing. The end face of the sensor is enclosed by a stainless steel net and the back side holds the connection terminals. Ethyl alcohol present in the breath is oxidized into acetic acid passing through the heat element. With the ethyl alcohol cascade on the tin dioxide sensing layer, the resistance decreases. By using the external load resistance the resistance variation is converted into a suitable voltage variation. The circuit diagram and the connection arrangement of an MQ 135 alcohol is shown below.
MQ-135 Circuit Diagram
A portable system has been developed to be used for safe driving applications. This device uses GSM mobile technology for the transmission of stress level to a remote location wirelessly. The advance interfacing of manmachine enhances the safety in driving.
1 H. Frumkin, L. Frank, R. Jackson, “Driving and Mental Health”, Urban Sprawl and Public Health: Designing, Planning, and Building for Healthy Communities, Island Press, 2004, pp. 140-142.
5 J.A. Healey, R.W. Picard, “Detecting Stress During Real-World Driving Tasks Using Physiological Sensors”, Journal IEEE Transactions on Intelligent Transportation Systems, Vol. 6 Issue 2, 2005, pp. 156-166.
6 J.F. Thayer, F. Ahs, M. Fredrikson, J.J. Sollers III, T.D. Wager, “A MetaAnalysis of Heart Rate Variability and Neuroimaging Studies: Implications for Heart Rate Variability as a Marker of Stress and Health” Review, Neuroscience and Biobehavioral Reviews, 36, 2012, pp. 747-756.
7 J.F. Thayer, R.D. Lane, “Claude Bernard and the Heart-Brain Connection: Further Elaboration of a Model of Neurovisceral Integration” Review, Neuroscience and Biobehavioral Reviews, 33, 2009, pp. 81-88.
8 A. Voss, R. Schroeder, M. Vallverdú, et al., ” Short-term vs. long-term heart rate variability in ischemic cardiomyopathy risk stratification.”, Frontiers in Physiology. 2013;4, pp. 364- 379.
9 L.I. Vanderlei, C.M. Pastre, R.A. Hoshi, T.D. Carvalho, M.F. Godoy,” Basic notions of heart rate variability and its clinical applicability”, Rev Bras Cir Cardiovasc 2009; 24(2), pp. 205-217.
10 Hoang ChuDuc*, Kien NguyenPhan, Dung NguyenViet, “A Review of Heart Rate Variability and its Applications”, APCBEE Procedia 7 (2013), pp. 80 – 85. 12 S. Boonnithi and S. Phongsuphap, “Comparison of heart rate variability measures for mental stress detection,” 2011 Computing in Cardiology, Hangzhou, 2011, pp. 85-88.
13 R. Costin, C. Rotariu and A. Pasarica, “Mental stress detection using heart rate variability and morphologic variability of EeG signals,” 2012 International Conference and Exposition on Electrical and Power Engineering, Iasi, 2012, pp. 591-59