Sensor Fusion

a prerequisite for intelligent light

The evolution from halogen to xenon to LED has vastly expanded the potential of vehicle lighting. Individual control of modern LED headlamps takes precision, light distribution and efficiency to new heights. In conjunction with intelligent control electronics, light becomes an assistant in all driving situations.

Adverse weather conditions, narrow lanes, twisting roads, bumps and dips all make life difficult for the driver, but adaptive, intelligent vehicle lighting can greatly reduce the burden and therefore improve road safety. In combination with safety and assistance systems, light is a prerequisite for reliable trajectory planning, micro navigation and detection of other road users. Powerful algorithms that allow data sources to be fused in the vehicle are crucial in this context.

With connected control, the sensors in vehicle cameras, distance meters and headlamps can be used to model the vehicle environment even more precisely and ensure that light and other assistance systems benefit from each other. Sensor fusion and connectivity are at the heart of OSRAM Continental’s approach to intelligent vehicle lighting.

Precise control

model-based development is one of our competencies

One advantage of modern light control units is their connectivity. Today, data transfer is already a prerequisite for tapping the full potential of the control elements. Adaptive functions – with and without fully automated assistance – operate with greater precision the more data is available and the faster it can be processed.

Our competence lies in the model-based development of connected control, which we test using algorithms developed in-house and validate in test vehicles. The model takes into account the vehicle, the driving dynamics and oncoming traffic while also processing relevant data from the on-board sensors and making use of navigation data and cloud-based information. 

The precision and responsiveness of the assistance systems are crucial factors for the performance of the lighting system.

Short latency between registering information and responding to it, and the weighting assigned to the available data pose a significant challenge.

We can embed the necessary lighting algorithms into the preferred architecture and partition them to different control units as required. We analyze both distributed and concentrated architectures to determine the best path forward for modular software development and interaction with the rest of the vehicle.

Fields of application

and our competencies

Intelligent lighting is the result of connectivity and system integration. The following four use cases reveal the potential:

Connected high beam assist

High beam assist is a good example of an application for intelligent data fusion. Many drivers change from low to high beam manually. With appropriate connectivity and back-end support, modern sensor systems can process and support this reaction. With the aid of cloud support, the vehicle can compare individual driving behavior with crowd sourcing data and assist with the appropriate action. This boosts convenience for the driver and improves safety for all other road users.

AR high beam stabilization

The technological possibilities of augmented reality also play a role in vehicle lighting. In the case of high beam stabilization, they improve safety for the driver and for the environment. For example, by processing distance data, the high beam can be more precisely aligned, to avoid glaring oncoming drivers. This is the key to making full use of the benefits of high-resolution headlamp systems. Full dynamic stabilization of light distribution enables symbols as warning signals or navigation guides to be placed in the real environment with high precision in order to create the effect of augmented reality.

Camera support

Intelligent light also supports other driving assistance systems. It can be used, for example, in poor visibility due to the weather. If rain or fog affect visibility, this will have an impact not only on the driver's reaction times but also on the accuracy of readings from various on-board systems. Light can support camera systems, close data gaps and maintain or even improve system functions by appropriate adaptive illumination of the road.

Structured light

Intelligent light can also help to provide 3D information about the vehicle’s environment. Modern HD light modules can deliver information on distances and road conditions by projecting light grids. In this way, they improve the precision of assistance systems and can act as a redundant channel. This can enhance general road safety, for example, improving the measurement of distances in front of the vehicle and detection of obstacles on the road surface.

Intelligent light as a forerunner for highly automated driving

The intelligence of light and light control will become increasingly relevant on the journey toward highly automated driving.

While comfort, convenience and safety have already been significantly enhanced by the networking of on-board data, information from local vehicle-2-vehicle communication and from connected cloud systems, such as weather reports or warning signals from vehicles up ahead, will offer added value. Smartphone data and functions will also become more relevant, especially in mixed traffic situations.

Light will the be key to highly automated driving, as it can provide situation-specific signals to drivers and the environment such as the transfer of control to the vehicle and the switch to a rest period. Another example is the switch from automatic to manual control, which could be indicated through a corresponding light signal.

This in turn will contribute to building trust in the technology. Vehicle light can therefore enable communication inside and outside of the vehicle. OSRAM Continental’s goal today and tomorrow is to provide drivers with the best lighting profile for their immediate situation.