“As ADAS technology expands into time-sensitive critical applications such as emergency braking, forward collision warning and avoidance, and blind spot detection, it can lead to a safer autonomous driving experience by combining data from multiple sensors to provide reliable decisions in real time .

“

As ADAS technology expands into time-sensitive critical applications such as emergency braking, forward collision warning and avoidance, and blind spot detection, it can lead to a safer autonomous driving experience by combining data from multiple sensors to provide reliable decisions in real time .

Using an AI-assisted camera that recognizes road signs and keeps the vehicle within the lane markings, it enables smarter and safer driving. But what happens when the camera’s view and the driver’s view are equally affected by fog?

“Cameras may be great at object recognition, but they’re not as effective in bad weather or at night,” said Miro Adzan, general manager of Advanced Driver Assistance Systems (ADAS) at Texas Instruments. Keep working. Driver assistance systems need to integrate many different sensors so that the vehicle can take full advantage of these different technologies while driving.”

With the advantages of different types of sensors, it is not only possible to switch between different conditions or applications, even in clear weather, the camera can more accurately capture the details of the object, and the radar will more accurately measure the distance of the object.

As the use of such systems expands into time-sensitive critical applications such as emergency braking, forward collision warning and avoidance, and blind spot detection, design engineers need to integrate these disparate sources of information into a single picture from which to make real-time reliable decision making.

“For autonomous parking, you need to combine data from cameras, radar, and sometimes ultrasonic data to get an accurate picture of what’s going on around the vehicle,” said Curt Moore, general manager of Texas Instruments’ Jacinto™ processors. Using these sensors alone None of the data produced is accurate enough, but when combined, you can get a more accurate picture of the space around you. That way, you can safely park in tighter spaces.”

Popularity of automotive sensors

Now, advanced safety systems are no longer just the preserve of high-end cars. Nearly 93 percent of vehicles produced in the U.S. are equipped with at least one ADAS feature. By September, 99 percent of new cars in the U.S. will have automatic emergency braking as standard. 1

This shift stems from the decreasing cost and size of sensors, such as TI’s mmWave radar sensor, which integrates the entire radar system into a coin-sized chip.

“Ten years ago, radar was mainly used for military applications due to size, cost and complexity. Today, radar is about to become standard in cars,” Miro said.

While the proliferation of affordable sensors has opened up new areas of application, it has also created new challenges for ADAS engineers. They need to design systems that aggregate and efficiently process all data streams while meeting stringent price/performance and power consumption constraints.

Communication challenges

In a single-sensor ADAS system, object detection data can be preprocessed near the sensor so that the information can be used immediately. However, sensor fusion technology requires immediate transmission of raw, high-resolution data to a central unit for processing, resulting in a single, accurate model of the environment that helps vehicles avoid collisions.

Heather Babcock, general manager of FPD-Link™ products at Texas Instruments, said: “Because all data is coming from these sensor nodes, it’s important to ensure that all data is synchronized so that the vehicle can sense what’s happening around you and make critical decisions. In order to transmit isochronous data in real-time, high-bandwidth, uncompressed transmission is necessary because compressed data creates delays.”

We created the FPD-Link communication protocol to transmit digital video streams from graphics processors to digital displays, and today is designed to transmit large amounts of uncompressed data over meters with simple, easy-to-route cables.

Heather said: “You have a standard protocol on this end and an FPD-Link serializer with a very secure and reliable proprietary encoding that converts the data stream. This is paired with a deserializer on the other end. To match, the deserializer reconstructs the data to its original format and transmits it over a variety of other interface protocols supported by TI’s portfolio.”

Enable more effective decision-making

Once this data reaches the central processing unit, often computationally intensive signal processing and deep learning algorithms are needed to integrate it into a unified model of the car’s surroundings, increasing the power input and heat output required.

Due to the physical limitations of the car, the size and weight of the battery and cooling functional infrastructure are strictly limited, so ADAS engineers need to design processors specifically to perform such tasks as efficiently as possible.

Texas Instruments’ Jacinto processors combine dedicated digital signal processing (DSP) and matrix multiply cores to operate at the industry’s lowest available power, even at temperatures as high as 125 degrees Celsius.

Curt said: “There are huge advantages to integrating the DSP and processor into a single system-on-chip. When used separately, they each require separate memory and power supplies, which increases system cost. Another advantage is that such The operation is integrated into a single chip, which reduces latency.”

In addition to energy-efficient processors, TI’s automotive-grade power management ICs feature functional safety features for sensor fusion, front-facing cameras, and domain controllers that improve the vehicle’s overall power efficiency and functionality.

Beyond a single device, TI’s entire ecosystem of ADAS products has been created for seamless compatibility, allowing automakers to choose from an overall product portfolio that can scale according to the needs and price points of their vehicles.

Miro said: “We take into account the various challenges faced by the vehicle in all ADAS blueprint design stages. This makes it easier for our customers to design the system.”