Next-Generation Head-Up Displays
HUD 2.0 may function as a primary information-display technology for vehicle drivers rather than a source of merely helpful or ancillary information.
by Alan Rankin and Jason Thompson
CURRENT head-up display (HUD) systems tend to display redundant information – that which is generally available elsewhere in the vehicle. HUD 2.0 – the next generation of this technology – is positioned to become the display of choice for Advanced Driver Assistance Systems (ADAS) information. With the addition of on-board sensors, cameras, and vehicle-to-vehicle/infrastructure communications, the amount of information a vehicle knows about its surroundings has increased exponentially. The challenge lies in how to effectively communicate what critical information is “known” by the vehicle and relay it to the driver. The need will only increase as we move toward semi-autonomous and autonomous driving capabilities. Unlike existing HUDs that tend to be used as second-ary displays in user-interface paradigms, HUD 2.0 may be central to the human–machine-interface (HMI) strategy and will function as a primary information display. As such, exceptional image quality and consistent readability in varying sunlight conditions are requirements.
A natural and intuitive way to communicate this information would be to use HUD 2.0 to augment the driver’s world-fixed view with conformal graphics that indicate what the car knows. Features such as navigation indicators, lane-departure warning (LDW), and adaptive cruise control (ACC) indicators could be displayed at a natural image distance as seen from the driver’s perspective. In Fig. 1, a HUD augments the reality of the driver’s view to provide useful information in real time. Note that the image appears in bright vivid color and is overlaid at the natural distance of the objects so that the driver can easily use the information with minimal distractions.
Fig. 1: An augmented-reality HUD shows conformal graphics over the driver’s line of sight.
Challenges
Although a detailed description of HUD design is beyond the scope of this article, we will review some of the key parameters. As shown in Fig. 2, both the field of view (FOV) and virtual-image distance (VID) play a role in determining perceived virtual-image size. While conventional HUDs cover only a fraction of a single lane, HUD 2.0, with a much larger FOV and longer VID, allows the driver to see images beyond a single lane of traffic. These increases in FOV and VID require higher luminance levels, more saturated colors, higher power efficiency, and increased tolerance to sunlight intensity. Additionally, these new parameters need to be achieved while also meeting all of the conventional automotive environmental conditions. Table 1 lists some of these parameters for HUD 2.0 as compared to a conventional HUD system.
Fig. 2: Both field of view (FOV) and virtual-image distance (VID) impact the perceived size of a HUD image.
Luminance and Power Efficiency
A larger FOV and higher luminance levels result in an easy-to-view image for the driver. To help ensure readability in varying sunlight conditions, the HUD should be capable of producing a virtual image between 15,000 and 30,000 cd/m2 to provide a proper contrast ratio over a wide range of roads and sunlight illumination conditions. A road covered by snow and directly illuminated by the sun would be the biggest challenge. However, the absolute power needed to create this image should remain low – both to minimize the volume needed for thermal management and also to keep the luminous flux in a workable range of the LED light source. To achieve both a larger FOV and higher luminance while not increasing power, a much more efficient imager is required. (For more about HUD luminance requirements, see the sidebar “HUD Legibility.”)
The Texas Instruments DLP 0.3-in. WVGA Type A100 digital micromirror device (DMD) is one possibility for integration into a HUD 2.0 system. It is >66% efficient and dramatically improves the system’s efficiency to enable the above parameters to be met. A HUD system based on DLP technology and RGB LEDs can achieve the required luminance and larger FOV. For example, a system designed with the 0.3-in. WVGA DMD and OSRAM Q8WP RGB LEDs2 uses only 6.0 W of LED power to achieve over 15,000 cd/m2 with an FOV of 10°, which is less power than even smaller secondary HUD systems today. The efficacy of this system (lumens per watt) is 10.6 lm/W.
Color Saturation
Many conventional TFT-LCD HUD designs use a white LED that is filtered into the component red, green, and blue colors. In contrast, HUD systems based on DLP technology use red, green, and blue LEDs and provide more saturated colors. This allows for increased readability of the image on the HUD display.3 Key performance metrics are used to judge the color performance of a system, including the gamut size measured by comparing its color gamut to the Rec. 709 color space, the hue of each color as defined by its dominant wavelength, and the saturation of the color.
Table 2 compares the TFT-LCD white-LED architecture1 with HUD architecture based on DLP technology with RGB LEDs. The RGB LED shows significantly higher performance both in the increased color gamut compared to Rec. 709 and the deeper saturated red and blue colors.
Sunlight Thermal Loading
As the FOV of a HUD system increases, so does the amount of sun energy collected by the HUD optics. Also, as the VID increases to allow the driver to view the image at the proper perspective relative to the real world-fixed view, the energy from the sunlight becomes more focused onto the internal imager of the HUD. The effect of both collecting more sunlight and focusing this energy to a smaller spot on the internal imager can be damaging to the imager due to the amount of heat collected in a small area. The HUD system based on DLP technology uses a diffusing screen material to create the internal image of the HUD system. For a conventional HUD system, the imager (typically a TFT panel) directly emanates the HUD image (Fig. 3).
Fig. 3: This block diagram shows the architecture for a conventional non-diffuser screen HUD design.
The diffusing screen is a passive element with two primary advantages: (1) it does not absorb the sun energy – it diffuses the light – and (2) it is not a source of heat itself. These attributes allow HUD systems based on DLP technology to more readily scale to the large FOV and longer VID needed for augmented-reality HUD systems. In addition to being bright enough to be seen in various ambient light conditions, a HUD virtual image should also be readable when the driver is wearing polarized sunglasses. Since DLP technology projects unpolarized light, this gives OEMs the ability to optimize the HUD for use with polarized sunglasses.
Environmental Conditions
Imaging technology used in automotive HUD systems must also be able to reliably operate in rigorous environmental conditions such as high humidity, extreme temperatures (including dramatic temperature changes), shock, and vibration. The DMD is a microelectro-mechanical system (MEMS), and some may wonder about its ability to meet the temperature cycle, shock, and vibration experienced in an automobile. The DLP 0.3-in. WVGA Type A100 DMD meets these conditions. Its mechanical structure is robust under shock and vibration in the <5-kHz range because the mirror resonant frequency is well above 100 kHz. Table 3 lists some of the critical tests that have been successfully completed on the 0.3-in. WVGA Type A100 DMD without issue.
Enabling the Next Generation
Automotive HUDs are becoming a more critical part of vehicle HMI strategies, especially as more and more ADAS technology is deployed in the vehicle. With HUDs transitioning from small secondary displays to large primary displays, the expectations for image quality, readability, and reliability increase. As mentioned earlier, HUD designers are particularly challenged by the need to achieve the above requirements while also meeting all of the conventional automotive environmental conditions.
Acknowledgment
Data throughout this article provided by TI unless otherwise noted.
References
1E. Buckley, “Pixtronix DMS™ technology for head-up displays,” Proc. SID Vehicles and Interfaces Conference (2011).
2OSRAM Opto Semiconductor. [Online]. Available: http://www.osram-os.com/osram_os/en/products.
3B. E. Blankenbach, “Comparison of the Readability of Colour Head-Up Displays Using LED and Laser Light Sources,” SID Symposium Digest of Technical Papers (2010). •
Fig. A: Dr. Larry Hornbeck received the 2014 Scientific and Technical Academy Award® of Merit (Oscar® statuette) for the invention of the digital micromirror device (DMD) technology as used in DLP Cinema® projection.