Safety system sensing technology improves driving comfort and convenience

Vehicle architecture is becoming more and more complex, and effective sensing elements are needed to improve the automation system, provide assistance to the driver and fully control the vehicle. Among the various types of sensors widely deployed , non-contact sensors are one of the most important sensors, which can realize many functions such as sensing the vehicle's environment and assisting in avoiding obstacles. However, the focus of most attention at present is on in-vehicle sensing for passengers for safety reasons.

Melexis  looks at the leading technologies used in this field and explores how to successfully improve driving safety, as well as enhance comfort and convenience.

In-car time-of-flight sensing principle

Time of Flight (ToF) sensing technology can detect objects and provide information about their location and movement in 3D space. It can also identify objects by detecting their shape, size, and orientation. The technology works by detecting reflected light from objects, using  LED  arrays or VCSEL devices combined with beamforming optics to produce a light source that illuminates the scene of use. Advanced sensing ICs are able to detect and measure reflected light from all objects within a certain range.

Figure 1: Block diagram of a typical ToF sensing system, including light source, sensor, and supporting chip

Light travels at a known speed, and measuring the time it takes for the reflected light to reach the sensor (the light's "time of flight") can be used to calculate the distance of an object. This method, called "direct TOF" (dTof), directly calculates time using a very precise time base and is suitable for longer distance applications such as LIDAR. dToF is generally considered a low-resolution technology that requires complex (and expensive) mechanical scanning to achieve high resolution.

Indirect ToF (iToF) calculates distance based on the phase shift of a known reference signal rather than absolute time. Such techniques are well suited for high-resolution applications, generating real-time 3D video at QVGA or higher resolutions using modern CMOS pixel arrays.

iToF's unique combination of medium/high resolution depth and amplitude (grayscale) mapping enables complex identification (such as people or objects) even when there is a lack of color contrast between the object and its surroundings. Based on these advantages, iToF overcomes the limitations of other technologies and is ideal for many forward-looking applications.

At a relatively close distance (about 5 meters), ToF can identify objects and free space, so the ADAS system can intelligently predict the possible movement of objects (if any) and take appropriate evasive action. Melexis' iToF is relatively unaffected by temperature and light changes, making it ideal for such exterior applications.

In-car applications using time-of-flight sensing

Improving road safety is one of the main drivers of the goal of fully autonomous vehicles. According to the latest report released by the National Highway Traffic Safety Administration (NHTSA), more than 90% of accidents are caused by driver error. If such accidents can be effectively eliminated, the roads will be safer. Although we cannot mass-produce fully autonomous vehicles in the short term, iToF can effectively and significantly improve road safety by monitoring drivers and their behavior.

Fatigue driving is a serious problem, and the high resolution achievable by iToF can determine if the driver is looking at the road ahead, yawning frequently, or even drowsy. Detecting such situations and suggesting (or even enforcing) the driver to take a break can effectively avoid potential accidents and save lives. For other driver behaviors such as not holding the steering wheel correctly, eating while driving, or using a mobile device in hand, the technology can also identify and issue warnings or take actions, and even stop the vehicle safely if necessary.

Airbags have saved countless lives and are now found in almost all cars. However, there are also cases where airbags cause injuries to infants or elderly people, or worse. ToF can detect the size of the passenger and estimate their weight to adjust the deployment of the airbag as needed. If there is no passenger in the seat, ToF can avoid unnecessary deployment of the airbag.

Some current hybrid vehicles will start and run the internal combustion engine to charge the battery when the battery is low . As traditional ignition keys are gradually eliminated, it is easy for drivers to leave the vehicle without turning off the engine, and the vehicle may start automatically without being attended. This will cause safety hazards, especially in confined spaces. However, ToF-based occupant detection can easily prevent such situations from happening.

In addition to improving vehicle safety, the ToF system can also provide a series of comfort and convenience experiences for the drivers and passengers in the car. For example, when the driver and passengers get in the car, the seat can move closer to the seat belt; the storage box will automatically light up when the passenger reaches out to use it; the operation of the in-car infotainment system can be automatically adjusted according to the number and position of the passengers in the car.

As vehicles become more sophisticated, the human-machine interface (HMI) in the cab becomes more complex. Using ToF sensors with projectors, control panels can be provided on a variety of available surfaces, greatly improving operational convenience and flexibility.

It is clear that ToF will help in-vehicle sensing applications achieve a leap forward, bringing higher precision to vehicles, thereby effectively improving safety and comfort.

Latest ToF Technology

Melexis' second-generation chipset includes a dedicated TOF sensor and a companion chip that can control the system and connect to an external microcontroller or serializer.

Figure 2: Key functional components of Melexis’ second-generation ToF solution

While retaining the same compact 5.5 mm x 6.5 mm package and optical format as previous generations, the new chipset doubles sensitivity and offers a choice of pixel-level gain for improved low-light performance. The system is aided by 940 nm illumination, which is invisible to the naked eye, greatly assisting nighttime operation inside the vehicle and taking advantage of lower daylight radiation at this wavelength.

The new sensor is more efficient, consumes 30% less power, generates less heat and saves energy costs. With improved signal-to-noise ratio, the system can increase the effective working distance by more than 65% at the same lighting level, or have lower lighting requirements at the same working distance as the previous generation. This feature can reduce the cost associated with light sources such as LEDs or VCSELs. It is now also possible to build dual cameras using a single companion chip.

Figure 3: Improved signal-to-noise ratio improves performance and reduces lighting requirements

Another new feature – pixel binning – combines four (2x2) or sixteen (4x4) pixels when lower resolution is required, reducing data throughput and supporting lower-cost host processors.

Summarize

In-cabin ToF sensing will help automakers meet upcoming regulations, including the NCAP 2025 proposal for occupant detection, while also improving the passenger experience in a meaningful way.

Over the past decade, Melexis has developed a variety of ToF solutions, and its second-generation solutions have higher performance, lower energy consumption and lower system cost, which can significantly reduce the application challenges faced by designers.