Challenges, opportunities & the future
The old axiom that “safety sells cars” has never held truer than it does today. As consumer awareness of advanced safety features continues to grow, so does regulatory efforts led by the global New Car Assessment Program (NCAPs).
In the evolution from L2-class vehicles to fully autonomous L5 vehicles, safety concerns are top-of-mind for consumers. Automotive OEMs and tier-one suppliers are racing to meet these expanding requirements while driving safety feature costs downward.
At the vehicle sensor level, this requires a careful price/performance balance among camera, RADAR and LiDAR technologies as no single technology is complete in its ability to enable fully autonomous driving or advanced driver assistance system (ADAS) functionality.
For the foreseeable future, RADAR and camera sensors will be the optimal sensor complement for vehicle safety applications enabling the next generation of safer, smarter cars. While camera sensors are essential for recognizing and classifying objects, RADAR sensors are vital for low light and challenging weather conditions. RADAR sensors also provide the vehicle-to-object distance, depth and velocity capabilities that cameras cannot.
Here are some of the trends, challenges and opportunities fuelling the industry’s continued innovation in high-resolution RADAR for advanced vehicle safety features.
HIGH-SPEED PROCESSING FOR HIGH-RESOLUTION RADAR
For automotive radar applications, the switch from the 24GHz to the 77GHz band well in motion. This expansion enables up to 5GHz of sweep bandwidth where previously 24GHz narrow-band RADARs were limited to 200MHz. This translates to 25x improvements in range resolution, allowing better ability to discern between multiple objects and distance between these apart objects.
- Distance to object
- Relative radial velocity
- Angle of arrival
- Radar cross-section (object size)
Suppliers continue to improve RADAR resolution however managing the trade-offs remains a challenge particularly with angular resolution, which distinguishes cars in neighboring lanes and is essential for adaptive cruise control (ACC) and automatic emergency braking (AEB).
Accurately gauging tunnels, bridges, signposts and attendant vehicle clearance leeway can have significant ramifications for detecting small obstacles. This new ability may enhance requirements for height detection and leverage sensors with 3D measurement (including angle measurement in azimuth and elevation). Manufacturers can achieve this goal by using the multiple input multiple output (MIMO) technique.
The benefit of this method is that it doesn’t add significant cost to the system, but it does achieve a significant improvement in angular resolution while minimizing the number of transmitted RADAR chirps.
RADAR has two major functional blocks:
RF Sensor (“Front End”)
- Antenna outputs
- TEF810x RF Sensor
- Signal creation and transmission
- Signal reception and signal conditioning
- Analog-to-digital sampling
Computational (RADAR MCU)
- Convert sampled signal into frequency information
- Identify “targets”
- Calculate 1) distance, 2) relative radial velocity and 3) angles of target
- Advanced functions such as classification and tracking
The processor performance demands of MIMO requires a highly capable processing platform with ample memory, bandwidth and signal processing capabilities to realize improvements in angle resolution.
PERFORMANCE AND POWER EFFICIENCY
Power efficiency is a critical consideration for every automotive sensor application. Power usage and thermal management grow evermore challenging as sensor networks increase in density, with fewer airflow/cooling outlets available. Therefore, increased RADAR requirements cannot be addressed by simply increasing the general-purpose processor cores beneath the sensor.
PERFORMANCE AND POWER EFFICIENCY
The RADAR sensor processing pipeline requires the utmost attention to performance-per-watt metrics, and it’s in this domain that NXP has established clear advantages compared to competing approaches.
NXP RADAR processors offer a scalable family of hardware and software-compatible products featuring S32R2x and S32R37 RADAR MCUs. These devices offer significant performance-per-watt improvements over traditional DSP1s by integrating a highly-efficient, specialized hardware accelerator — NXP’s proprietary Signal Processing Toolbox (SPT).
S32R2x and S32R37 are 32-bit Power Architecture® based MCUs designed to address advanced RADAR signal processing capabilities enabling longer range, higher resolution and accuracy for RADAR-driven, safety-critical applications.
NXP S32R2x Additional Capabilities Long-range RADAR, including forward-looking sensor, adaptive cruise control (ACC), autonomous emergency braking (AEB),rear-traffic-crossing alert (RTCA) and pedestrian protection.
SPT supports high-performance RADAR signal processing using engine integrating signal processing operations, to provide the required hardware modules to help engineers with:
- ADC sampling within a programmed capture window
- Hardware acceleration for fast Fourier transform (FFT), histogram calculation, 2D peak search and mathematical operations on vector data
- High-level commands for signal processing operations
- Compression/decompression for DMA data transfers
- CPU interrupt notification and a watchdog
NXP Solution NXP provides cutting edge technology in a scalable full system solution across models as well the ability to scale into the future as innovation continues to advance. Our full solution offering, from transceiver, processor, and power management integrated circuits (PMIC) to the safe and secure networking on which it relies, means we can go beyond simply delivering components to supporting the full radar application.
THE ROAD AHEAD
Where driver, passenger and pedestrian safety measures are concerned, every millisecond of vehicle response time is critical. As demand for advanced safety features and higher resolution RADAR sensors increases, processing platforms will need to balance compute agility and power efficiency.
NXP AND THE FUTURE
NXP’s comprehensive portfolio of RADAR MCUs – in combination with its industry-leading SPT hardware acceleration capability – includes a scalable, streamlined and highly integrated solution designed to enhance RADAR resolution and sensor data fusion for the next generation of safer, smarter vehicles. NXP will continue to keep pace with growing L2+ requirements, enabling tier one suppliers to build the highest performance RADAR solutions in the industry.