• This email address is being protected from spambots. You need JavaScript enabled to view it.

Trends In Automotive Radar

Autonomous Vehicle

Development Flow Overview

Automotive Radar Overview

Radars are used as part of the sensors that provide information about the enviroment in an autonomous vehicle. Automotive radar are generally classified in three categories: SRR or Short-Range Radar, MRR or Mid-Range Radar and LRR or Long-Range Radar.

SRR works between 1 and 20 m from a vehicle. They are typically used to detect possible dangerous situations near the vehicle. SRRs are normally included in functions like blind spot detection, parking aid, lane change aid, side impact and collision avoidance.

MRR are used in the same type of applications as the SRR, but with a range of 1 to 60 m. The biggest difference between SRR and MRR, apart from the distance would be the operating frequency, where SRR typically work in the 24 GHz spectrum and MRR in the 77/79 GHz spectrum. As of 2022 Europewill ban the use of wide-band operation in the 24 GHz spectrum, and thus limit some of the use of the radars operating at this frequency. Other benefits with the 77/79 GHz spectrum is the lower form factor, where antennas would be one third of the size of the 24 GHz antennas. It also has a wider frequency band, and high-resolution radar implementations are based on this frequency spectrum.

LRR typically operate on distances up to 250-300 m. They are used to enable automated driving in applications like adaptive cruise controls, collision avoidance and autonomous emergency braking.

Cost Optimized

Image

The cost optimized radar consist of a single board, containing the antenna, RF, Power Management and a processing unit. The processing unit will handle the signal processing required for the radar signals (chirp FFT, clustering, Kalman filters, RQ & Cholesky decomposition, SVD of non-square matrix, Capon matrix extraction, MUSIC eigenvector extraction etc.)

Typically this board will also contain some element of higher level processing, such that the data provided to the system will become a list of tracked objects. This may entail distance transforms, clustering, particle filtering and object detection and classification.

High Performance Radar

Image

A high-performance automotive radar system will consist of multiple radar modules, each with its own subsytem containing antenna, RF and power management. The main difference between the signle board and the high performance system lies with fact that the processing performed on the radar module for the high performance radar is limited to the radar processing (as described above). Any type of AI functionality, fusion and decision making is handled higher up in the system. This system will put high requirements on the ability to transfer data to the higher system, and for this system to be able to process the data in a timely manner.