Click to enlarge
Safety systems have come a long way from the humble seat belt driven partly by the need to make travel on the
road safer, but also by a stream of regulation ranging from mandatory Electronic Stability Control in the US
market to pedestrian safety measures in Europe that have fundamentally changed the design of the front end of
cars.
Today every OEM can achieve 5 star safety ratings if they have to and they are very adept at ensuring designs meet the necessary criteria whether in the Australian, US or European NCAP system. However, in the future this task is set to become more difficult as the achievements standards begin to reflect the increasing availability of active safety systems.
Active safety systems are those systems that act to warn or assist drivers in avoiding accidents, and they are increasingly the focus of the majority of safety development across the automotive industry. All active safety systems employ sensors and electronic data processing to monitor and respond to vehicle dynamics and driver behaviour. Advanced systems send signals, including IR (Infrared), LIDAR (Light Detection And Ranging) and radar, and use cameras and other sensors to receive reflected signals and process the data received electronically.
Some active safety systems, such as Lane Departure Warning (LDW) simply warn the driver, with sound or haptic (vibration) signals, if the vehicle begins to depart from its lane when the turn signal switch has not been used. The more advanced lane departure systems, Lane Keeping Assistant (LKA) systems, will actively intervene to turn the vehicle back into its lane with the movement of the steering wheel providing a further warning to the driver. However, driver input will always override the LKA system's steering interventions.
Currently, active safety systems can be considered in terms of those that operate by influencing vehicle dynamics and those that are essentially driver warning systems or systems that interact with the driver. However, as systems become more integrated, this delineation becomes less meaningful and at some future stage the integration of stability systems with external sensors and Adaptive Cruise Control (ACC) will mean that active safety and driver assistance systems become part of the same integrated safety system. Some suppliers and OEMs are referring to this integrated technology as Collision Mitigation Systems.
ABS systems were first marketed by Bosch in 1978 and are now standard fitment on almost all vehicles. As the name suggests, Anti-lock Braking Systems operate to prevent the braking system from locking the vehicle's wheels in order to maintain adhesion to the road surface and provide optimal braking. ABS sensors and components have become the foundation for the development of more complex systems, such as Electronic Traction Control (ASR) and ESC.
ASR, or Traction Control, activates when sensors detect that the vehicle is losing traction at one or more wheels and applies braking to the wheel(s) concerned in order to maintain tire adhesion. The ASR system typically employs additional sensors to those in the ABS system and provides the enabling technology for ESC systems. It is also possible to utilise an electronic differential locking system as a component of ASR.
ASR is widely available and is standard in most markets on luxury and sport segment vehicles as a way of managing increased engine power outputs.
Brake Assist systems assist in emergency braking by increasing the brake pressure, independently of how hard the brake pedal is pressed, when the pedal is applied suddenly. The system employs the ABS or EBS system and applies the pressure required to achieve optimal braking performance under the prevailing road conditions.
A further development then leads to predictive Brake Assist systems that use sensors to detect the sudden approach of the driver's foot towards the brake pedal and prime the pressure in the braking system so that the brake pads are in touch with the rotors. This reduces the brake system reaction time, saving valuable milliseconds in emergencies. Predictive systems can be utilised as a component of Active Cruise Control (ACC) to prime the braking system if the ACC system detects a possible emergency.
Click to enlarge
ESC/RSC systems interpret information from vehicle-based sensors, including, in some cases, roll sensors and
automatically apply combinations of engine torque and braking to stabilise yaw, pitch and roll. The systems are
intended to improve the stability of the vehicle in situations in which the driver has lost control, and are
particularly effective in reducing accidents on curves, on surfaces with limited adhesion and in vehicles, such
as SUVs, that have a high centre of gravity.
In the US, NHTSA reports have found that ESC/RSC systems are effective in reducing single vehicle accidents, such as rollover and unintentionally leaving the road. Manufacturers have already voluntarily fitted them to many SUVs and in the US fitment to all light vehicles is standard September 2011.
The future development of ESC/RSC systems has already been mapped well ahead, with high levels of integration predicted with passive safety systems and pre-accident systems such as adaptive cruise control (ACC).
Tire Pressure Monitoring Systems alert the driver if a tire loses pressure and can be extended to include the detection of other tire defects.
Research in the US has concluded that 53,000 accidents and 540 fatalities annually can be directly attributed to incorrect tire pressures. Similarly, in Germany, some 41% of all road accidents in which people are injured are due to incorrect tire pressures. In a recent study in Germany, it was found that 75% of vehicles had one or more tires under-inflated by at least 0.2 bar (2.9 psi).
Therefore, not only is the fitment of TPMS likely to have a real effect on safety, with the growth in the market underwritten by legislation, at least in the US, there is a compelling commercial case for suppliers in the sector - in marked contrast to many other active safety systems, where the commercial case for development can be somewhat tenuous.
Recent studies have indicated that driver fatigue and inattention are major causes of fatal road accidents. A significant number of these accidents could be prevented if the vehicles were equipped with Drowsiness Detection Systems (DDS). Facial recognition technology is already well developed, and provides the basis for creating a baseline of the driver's facial features. Blink responses, head position and head orientation are then monitored and software algorithms employed to measure fatigue and inattention responses against the baseline. To compensate for the wide variation in natural light conditions, DDS uses IR technology to illuminate the driver's face in a non-intrusive way. The use of more than one camera can even overcome reflection problems from glasses and sunglasses, by virtue of the reflected image having different locations in the different images.
When the DDS determines that the driver is drowsy or inattentive, it activates warnings that can be audible sounds, haptic (vibrations) warnings through the steering wheel, pedals or seat, or combinations of these. DDS can be developed further so that, in the case of sustained driver drowsiness or inattention, the braking system can be deployed to bring the vehicle to a stop. Future integration with lane keeping systems and CMAS would enable the vehicle to be stopped as safely as possible.
Curve Adaptive Lighting (CAL) systems direct part of the light beam in the direction in which the vehicle is being steered. The degree to which the light beam is swivelled is governed by the degree of steering deflection and can be speed-sensitive, so that the deflection occurs more slowly at higher speeds. CAL is particularly effective on winding roads with low levels of or no street lighting, and is useful in helping to reduce rural accidents at night that involve pedestrians or other obstacles on the road.
CAL systems can also be developed to include the diverting of light beams to road edges and the lowering of the intensity of straight-ahead beams in situations where visibility is reduced, such as in fog, rain or snow, or in well-lit urban situations where increased lateral light can improve the identification of pedestrians or bicycles. CAL can also be adapted to raise low beams to increase visibility at higher speeds.
Whereas passive Vision Enhancement (VE) systems use a far-infrared (FIR) detector to sense thermal radiation from the scene in front of the car, active systems operate in the near infrared (NIR) and use an infrared source to illuminate the road ahead and standard silicon cameras (Charge-Coupled Device [CCD] or Complementary Metal Oxide Semiconductor [CMOS]) to generate an image of the IR-illuminated scene ahead.
Hella's active night vision system.
Click to enlarge
Because active VE systems are illuminating the scene, far more detail is visible in the image, including lane
markings and road boundaries, enabling the driver to differentiate between obstacles that are in their driving
lane and those that are not. As with passive VE systems, existing in-car display screens can be used for active
VE display and the system can also be used for other vision-based safety functions such as Lane Departure
Warning.
The IR illumination technology is most likely to be semiconductor laser or LED, the latter of which would be more cost-effective, although currently LED IR sources tend to be much bulkier than laser sources.
It is likely that the future development of VE systems will include the integration of inputs from an array of sensors that will be processed through complex algorithms in order to detect changes in road surface conditions, including the type of surface, and hence its grip characteristics and whether there is water or ice on the surface. Vision analysis systems, such as those already developed for proximity and blind spot monitoring, which track changes in images to assess potential hazards, are likely to be used in applications such as Adaptive Cruise Control. When these advanced systems determine the existence of a hazard, the driver will be warned via light displays in and/or sounds from the instrument cluster.
Click to enlarge
Lane Departure Warning (LDW) systems track road features, such as road marking or road edges, to determine
whether a vehicle is leaving the lane it has been in. Warnings are delivered via vibrations in the steering
wheel or seat, or through auditory signals such as a simple audible alert tone or the sound of driving over a
rumble strip - a sound that many drivers already recognise as a warning that the vehicle is crossing a road
marking line.
Provided the road has distinct features such as white lines to mark lanes and road edges, LDW systems are effective in preventing deviation from a chosen lane, or the road itself, due to driver fatigue, loss of concentration or poor driver lane discipline. The system does not deliver a warning if the driver has signalled a lane change, so it has the added safety benefit of reminding drivers to give lane change signals.
LDW systems integrate an array of technology, including external sensors (camera, radar or IR), steering angle sensors, wheel speed sensors and image processing software algorithms. In some cases, LDW systems also use a camera to monitor driver eye movements and determine whether a lane change is intended. LDW can be adapted for use in blind spot and proximity warning systems.
Click to enlarge
Lane Keeping Assistance (LKA) is a more advanced development of LDW. When an unintended lane change is
detected by the system, it intervenes in the steering to inhibit the lane change. The sensing and processing
technology is the same as that in LDW and auditory or vibration warnings can be employed along with the
steering input. Although the system gently moves the steering wheel in the direction required to stay in
lane, this is intended primarily as a signal to the driver, and the driver's input always takes priority.
LKA is effective in preventing unintended lane changes, both by alerting the driver that an unintended lane
change is occurring, and by keeping the vehicle in lane until the driver decides otherwise.
Adaptive Cruise Control (ACC) systems are radar- or LIDAR-controlled cruise control systems that maintain and manage vehicle speed in response to constant monitoring of the distance between the vehicle and any vehicle immediately in front. The systems send out signals that are reflected off the vehicle in front and detected by external sensors. Software algorithms then detect changes in the inter-vehicle distance and utilise the accelerator or the brakes to maintain a distance that is appropriate for the current speed.
Current versions of ACC, which were launched a few years ago include Low Speed Following (LSF) and 'follow-to-stop' functions are now becoming available that will enable ACC to be used in slow moving traffic. 'Follow-to-stop' actually stops the vehicle if the one in front stops. This, in turn will be extended to include a 'stop-go' function that will allow the vehicle to move off again once the one in front has moved off.
Further development is envisaged that could include communications between a GPS system, a digital roadmap and roadside signalling equipment that could be installed on urban roads and busy highways. Alternatively, in-vehicle telematics systems could, via local communications systems infrastructure, constantly create and modify Local Area Networks (LAN) in which vehicles in the same vicinity provide information to each other regarding vehicles that have slowed or stopped up ahead.
It is expected that ACC will provide the basic platform for Collision Mitigating Systems (CMS) that detect obstacles ahead and slow, stop or steer the vehicle accordingly.
Collision Mitigation Systems (CMS) use radar or LIDAR signals and sensors to detect imminent collision and then deploy interventions to mitigate the effects of the collision. For example, the system can activate the vehicle's brakes to reduce accident severity. It can be integrated with passive safety systems to prepare occupants for a collision by deploying seatbelt tensioners, closing windows and activating airbags before the collision occurs, potentially saving 20 to 50 milliseconds in airbag deployment time. Some systems monitor the rear of the vehicle and can adjust head restraints to minimise whiplash injuries. CMS can also be used to deploy systems such as bonnet-raising devices that are designed to minimise injuries to pedestrians.
CMS requires considerable systems integration, advanced vehicle electronics architecture, the integration of various sensor arrays and complex initiation strategies for the inter-related actuation of the vehicle's active and passive safety systems. There are significant issues regarding the prevention of the systems from calculating 'false positives' from stationary roadside objects and on-coming traffic and activating the vehicle's active and passive safety features as if there were an emergency.
The first commercially available CMS system has been launched by Lexus.
CMAS are currently under development and will use the same signal, sensor and electronic processing systems as CMS, but will extend the array of mitigating interventions to include a greater array of vehicle dynamics systems including accelerator, suspension, steering and AWD. As with CMS, CMAS will also be integrated with passive safety systems to minimise injuries to both vehicle occupants and pedestrians.
The extension of CMS to CMAS and its ability to use the steering, suspension and AWD systems further complicates the electronic architecture, processing software requirements and the integration of more vehicle control systems.
