Modern Radar for Automotive Applications
2: Department of Electrical & Computer Engineering, Texas Tech University, USA
3: Radar Technology Department, Netherlands Organisation for Applied Scientific Research (TNO), The Netherlands
Radar is a key technology in the safety system of a modern vehicle. Automotive radars are the critical sensors in advanced driver-assistance systems, which are used in adaptive cruise control, collision avoidance, blind spot detection, lane change assistance, and parking assistance.
The book covers all the modern radars used in automotive technology. A long-range radar mounted in the front of the vehicle is usually for adaptive cruise control. The medium range radars mounted in the front and rear provide wider coverage than the long-range radars and they can be used for cross traffic alert and lane change assistance. The corner mounted short range radars support parking aid, obstacle/pedestrian detection and blind spot monitoring. In real applications, these radars usually work together to provide more robust detection results. In this book, we also recognize that the future of automotive radars should not only address conventional exterior applications, but also play important roles for interior applications, such as gesture sensing for human-vehicle interaction and driver/passenger vital signs and presence monitoring.
The book is aimed at those radar engineers who are working on automotive applications.
Inspec keywords: radar antennas; road vehicle radar; radar signal processing; radar receivers; CW radar; sensors; millimetre wave radar; antenna arrays; radar transmitters
Other keywords: radar antennas; radar transmitters; radar receivers; automotive applications; antenna arrays; CW radar; modern radar; sensors; radar signal processing; road vehicle radar; millimetre wave radar
Subjects: Sensing devices and transducers; General electrical engineering topics; Radar equipment, systems and applications; Signal processing and detection; Antenna arrays
- Book DOI: 10.1049/SBRA553E
- Chapter DOI: 10.1049/SBRA553E
- ISBN: 9781839534355
- e-ISBN: 9781839534362
- Page count: 325
- Format: PDF
-
Front Matter
- + Show details - Hide details
-
p.
(1)
-
1 Introduction
- + Show details - Hide details
-
p.
1
–5
(5)
The term "radar" stands for radio detection and ranging. A radar is an electromagnetic system used to detect, locate, track, and identify different objects within a certain area. A radar transmits electromagnetic energy in the direction of targets to observe the echoes from them. The targets could be ships, aircraft, astronomical bodies, automotive vehicles, etc. In early days, radar systems were only used in the military area due to their bulky size and high cost. Thanks to the advance of high-frequency integrated circuit (ICs) and monolithic microwave ICs, modern miniature radar systems can be realized on a printed circuit board or even on an IC [1-5]. Applications of radar systems have been extended to commercial areas, such as through-wall detection [6-9], indoor localization [10-13], biomedical applications [14, 15], and driver assistance [16, 17]. The topic of this book focuses on radar technology and its applications on automotive vehicles.
-
2 Principles of automotive radar systems
- + Show details - Hide details
-
p.
7
–56
(50)
A radar system radiates electromagnetic energy into space from an antenna or an antenna array. The radiated electromagnetic energy "illuminates" the surrounding targets. The "illuminated" targets intercept some of the radiated energy and reflect a portion back to the radar system. The radar system utilizes one or multiple receiver channels to detect the reflected energy to determine the targets' range, velocity, and relative angles.
Based on the different types of waveforms radiated by the radar transmitter, radar systems can be categorized as pulsed radars and continuous-wave (CW) radars. A pulsed radar consists of a repetitive train of short-duration pulses. The range of the target is measured based on the time delay between the transmitted pulse and the received pulse. Different from a pulsed radar, a CW radar usually transmits the electromagnetic wave continuously in a period of time. The properties of the targets are obtained by comparing between the received signal with a replica of the transmitted signal. For automotive applications, CW radar systems have been the dominant due to their advantages in multiple aspects. Compared with a pulsed radar, a CW radar features low peak transmit power, simpler, and highly integrated structure, which make its applications spread into various areas, especially for automotive applications. This chapter is trying to present a thorough and consistent description of the fundamentals of radar technology for automotive applications. Though many of the concepts are the same between pulsed radars and CW radars, CW radars are emphasized over pulsed radars in this book.
-
3 MIMO radar technology
- + Show details - Hide details
-
p.
57
–94
(38)
Automotive radar with a small number of antennas has been used for advanced driver-assistance system (ADAS) purposes since the late 1990s. These early automotive radars mostly provided target detection and velocity information. However, the current generation of automotive radar for ADAS has rather limited ability to resolve closely spaced targets. LiDAR systems have a better angular resolution (less than 1 degree) and have been introduced in Level 4 and Level 5 autonomous driving systems. LiDAR can provide point clouds. Via the use of deep neural networks, such as PointNet [1] and PointNet++ [2], the point clouds can lead to target identification. However, due to its use of light spectrum wavelength, LiDAR is susceptible to bad weather conditions, such as fog, rain, snow, and dust in the air. In addition, the cost of LiDAR is high. On the other hand, automotive radar with millimeter-waveform technology has the potential to provide point clouds at a much lower cost than LiDAR and with more robustness to weather conditions. Such radar is referred to as a "high end radar" or imaging radar [3]. Computer vision techniques [1, 2] that were previously reserved for high-resolution camera sensors and LiDAR systems can be applied to imaging radar data to identify targets. For example, a car can be identified based on two-dimensional (2D) radar points of an imaging radar using PointNet [4]. Imaging radars have been attracting the interest of those developing fully autonomous vehicles, major Tier-1 suppliers, and automotive radar startups.
In addition to sensitivity, the important requirements for automotive radar are high resolution, low hardware cost, and small size. Multiple-input multiple-output (MIMO) radar technology has been receiving considerable attention from the automotive radar community because it can achieve high angular resolution with relatively small numbers of antennas and receivers. For that ability, it has been exploited in current generation automotive radar for ADAS as well as in next-generation high-resolution imaging radar for autonomous driving. For autonomous driving, information in both azimuth and elevation is crucial. In particular, the height information of targets is required to enable drive-over and drive-under functions. Two typical scenarios are shown in Figure 3.1. It is safe to drive over a metal beverage can on the road and to drive under a steel pedestrian bridge over the road. To meet such requirement, the array is required to have a large aperture in both azimuth and elevation. The MIMO radar is a good candidate for high-resolution imaging radar for autonomous driving. In the MIMO radar, the targets are first distinguished in range and Doppler domains. Then, large virtual arrays with hundreds of elements can be synthesized to provide high resolution in both azimuth and elevation. As a result, point clouds with similar performance to LiDAR can be generated at a much lower cost.
In this chapter, we introduce the concept of imaging radars using MIMO technology, present some examples for synthesizing hundreds of virtual array elements by cascading multiple radar transceivers with each supporting a small number of antennas, and discuss design challenges.
-
4 Interference and interference mitigation
- + Show details - Hide details
-
p.
95
–128
(34)
Recent improvements in self-driving technology emphasize the importance of sensors that are being used in vehicles. Achievements in integrated circuits and the semiconductor industry made the low-cost mass production of single-chip automotive radars possible. While other sensors exist, automotive radar acts as the digital eyes of self-driving vehicles due to its proven all-weather, day, and night capabilities, which makes the automotive radar one of the key elements for self-driving technology. Most of today's vehicles are already equipped with radar systems to improve situational awareness and road safety. Current mmWave automotive radar sensors share a spectrum space from 76 to 81 GHz [1, 2]. The increasing number of radar-equipped vehicles on the roads has already been an issue for the coexistence of multiple automotive radars in congested traffic because the unwanted radar signals generated by other radar - also known as interference - negatively affect the functionality of the radar systems by decreasing their sensing capability. Since a lot of equal or similar waveforms and transmission strategies are presently used in automotive radar applications, interference occurs between multiple radar units. This kind of interference may raise the noise floor and reduce the signal-to-noise ratio which degrades the probability of target detection. On the other hand, interference generated by other radar systems may cause (ghost) false targets that reduce the target tracking ability of the radar.
Recent automotive radar systems are taking advantage of multiple-input multiple-output (MIMO) antenna arrays to provide the azimuth information of targets. Depending on the MIMO antenna configuration, it is also possible to exploit the azimuth and elevation information of the targets. While interference is mostly generated from other radars, there might be self-interference from the strong return signal reflected by the radome or induced by the mutual-coupling (spill-over) effect between transmitters and receivers. It is crucial to mitigate interference or reduce its effect because the objects with low radar cross sections (RCSs), such as pedestrians or cyclists, may not be detected or be completely lost during tracking. Therefore, interference leads to dangerous situations and becomes a bottleneck for driving assistance and autonomous vehicles. Especially in fully autonomous vehicles, where any human intervention will no longer be present, the dependability on the sensors is extremely high and there is no tolerance for sensing failures due to interference.
In this chapter, we examine different types of automotive radar interference, their characteristics, and their effects on radar system performance, as well as provide a review of the current state of the art for interference mitigation techniques.
-
5 mmWave radar tracking and sensor fusion with camera
- + Show details - Hide details
-
p.
129
–159
(31)
W-band millimeter wave (mmWave) radar is becoming an essential sensor in advanced driver assistance system (ADAS) and autonomous driving fields [1-5], including adaptive cruise control, pedestrian detection, collision avoidance, lane changing monitoring, and emergency braking. ADAS systems comprise of multiple sensors, such as radar, camera, and LiDAR. Each type of sensor offers its advantage, e.g., radar has a reliably moving target detection and is robust to adverse weather conditions. But radar has a relatively poor angular resolution comparing with its optical counterparts, which can achieve 0.1-1 degree resolution in both azimuth and elevation directions. The shortcoming of radar will result in uncertainty, inaccuracy, and missing or ghost detections at certain distances. One way to reduce the uncertainties and failures is to fuse radar outputs with heterogeneous sensors [4, 6].
mmWave radar research has gained immense popularity with the recent increasing demand of automotive radars in ADAS and autonomous driving industry. Among all the researches on mmWave radar applications, tracking [7-9], target recognition [10-12], and sensor fusion [13-15] are the most popular topics. In this chapter, we will start with a reliable radar tracking method using extended Kalman filter (EKF) and discuss the nonlinearities of radar tracker. Then, we will further develop a radar-camera sensor fusion by constructing a fusion-extended Kalman filter (fusion-EKF) in the remaining of the chapter.
Sensor fusion has been developing rapidly in recent years. However, there are limited studies discussing fusion of mmWave radars with other sensors since radars provide a limited number of detection points representing targets-of-interest [4], which make it difficult to recognize from a snapshot of radar detection. But if fusion can be achieved before the target classification of radar, a fusion system may fully take the advantage of radar, i.e., increasing the reliability of detecting moving targets, avoiding blockage, and tracking dramatically. In this chapter, we aim to increase mmWave radar's informative capability about targets and further its versatility by fusion with monocameras as an example.
Specifically, we introduce a fusion-EKF, which is designed to fuse data from heterogeneous sensors such as mmWave radar and monocamera with real-time fusion algorithm running for tracking. Sensor fusion and association are done within the fusion-EKF using a homography estimation method (HEM) [16], timeline alignment, and region search. Reliable detection and cross-validated target tracking are also realized. As we do not use any machine learning-based approaches to realize the fusion, the introduced fusion method achieves a low computational complexity, which is ideal for implementing in real-time systems. The experimental results also show that the proposed system can provide a reliable tracking and detecting result with low calculation costs. An embedded system like Arduino or Raspberry Pi can be utilized to process the data for real-time applications.
For the new fusion-EKF, a new concept is introduced, i.e., error bounds (EBs), which is defined as the sensor's region of approximation. EBs are not from the uncertainty of sensors [17] but the sensors' resolutions from their respective perspectives. An HEM is applied in associating heterogeneous sensors via their EBs. The fusion-EKF is designed to take both radar and camera as inputs and associate the data inside the filter to obtain ideal target tracking outputs. Data association of the fusion-EKF is capable to support tracking of multiple targets.
The structure of this chapter is as follows. In section 5.2, related work on mmWave radar tracking and sensor fusion are studied and presented. In section 5.3, radar-EKF tracking methodology is shown. In section 5.4, the sensor fusion fusion-EKF is presented with preprocessing, data association and sensor synchronization. In section 5.5, results of radar-camera fusion-EKF and root mean square errors (RMSEs) of improved EBs are shown.
-
6 Automotive radar target classification
- + Show details - Hide details
-
p.
161
–193
(33)
As an emerging application of the modern radar technology, the automotive radar target classification is demonstrated in this chapter. This is a cross-disciplinary research that involved ML and radar scene simulation. Some commonly used ML models including MLP, CNN, and RNN (LSTM) are introduced. Besides these classification algorithms, a good classification model requires the differentiable input data as well. Depending on the types of radar and signal processing approaches, the radar data can be represented in many forms. Radar can provide excellent velocity detections, and some examples utilizing the micro-Doppler signature for dynamic target classification are provided. This approach shows good performance for distinguishing pedestrians from vehicles. In a more general scenario regardless the movement of targets, the classification models based on the statistical information of RCS are demonstrated. The models are trained with a large high-fidelity simulation dataset and are validated by measurement. Though the RCS-based classification models can be applied to both dynamic and static scenarios, for the targets in both near and far range, their performance is limited by the small number of features. For radars with imaging capabilities, the 2D or 3D radar images can be generated and utilized in the radar target classification. Excellent performance can be achieved, but the model based on radar images may only be applied very successfully only for targets that are not too far from the radar due to the limited angular resolution of the radar.
It should be pointed out that not all models or radar data for target classification are covered in this chapter. Also, some technical challenges related to the target classification are not discussed, for example, the algorithms to isolate and clean up the radar data for each target are of such method not discussed here. Although not all aspects of radar target classification are covered in this chapter, it is hoped that the reader is able to gain some sense of the available methods for radar target classification with ML.
-
7 Road condition recognition with radar
- + Show details - Hide details
-
p.
195
–225
(31)
Correct recognition of the road conditions is critical for driving safety. Identifying the road conditions with radar data is attractive since first radar can detect the road in front of the vehicle to provide some reaction time, and second, the mmWave radar is a robust sensor that can work under inclement weather conditions.
The radar signals scattering from the road surface have two contributors: surface scattering and volumetric scattering. In this chapter, the surface scattering at 77 GHz is studied with the full-wave simulation approaches, and the backscattering coefficients are modeled as empirical functions of the surface roughness, dielectric constant as well as the incident angle. The volumetric scattering is modeled with the radiative transfer theory and a semi-empirical model has been developed for the dry asphalt and concrete road. Moreover, the measured backscattering coefficients for different road conditions and road types are presented and analyzed, which shows that the polarimetric backscattering coefficients are promising to be applied to identify the dry, wet, icy and snowy road conditions.
There are still some limitations and the need for significant future work before radar data can be utilized for road recognition. First, the detectable range of the road condition is limited by the height of automotive radar, for example, even for a 2 m height radar with incident angle of 80°, its detectable range is only about 11 m. Second, as the backscattering power from the road is very weak, high gain antennas and large transmitting power are required, but they are usually limited by the radar design and not available for most automotive radars on the market. The measured road conditions presented in this chapter may not represent all possible cases that the vehicles may experience, for example, the road condition with heavy rain is not considered. In order to generate a comprehensive and robust road recognition model, more measurements for all kinds of road conditions are desirable.
-
8 Radar-based gesture sensing
- + Show details - Hide details
-
p.
227
–243
(17)
This chapter presents a basic introduction to short-range radar and types of gesture radar. The chapter is mainly divided into three parts, including continuous wave (CW) radar, frequency-modulated CW (FMCW) radar, and in the Internet of Things (IoT) era several application examples of short-range radar-based motion sensing in gesture recognition. The simple theory and signal processing algorithm of the radar is also briefly introduced. With the advantage of the small form factor, low power consumption, and low cost, we can envision the ubiquitous applications of short-range radar technology in the coming IoT era. This chapter introduces their potential application in automotive, mobile and wearables, smart home hardware, augmented reality/virtual reality, gaming consoles, metaverse, healthcare, and many other exciting applications and use cases.
-
9 In-cabin vital sign monitoring
- + Show details - Hide details
-
p.
245
–302
(58)
As automobile craftsmanship evolves, automobiles now include intelligent technology, environmentally friendly manufacturing process, luxurious interior, and comfort, other than basic transportation needs. These additions provide much better driving and passenger experiences. Nowadays, the number of vehicles per capita is significantly increased, safety, however, should always be the priority for every car, and the definition of safety covers drivers, passengers, freight, pedestrians, and other vehicles.
-
10 Conclusion
- + Show details - Hide details
-
p.
303
–305
(3)
Thanks to the advances of embedded signal processing, high-frequency integrated circuit (IC), and monolithic microwave IC, miniature radar systems can be realized on a printed circuit board or even on an IC [1-5]. Applications of radar systems have been extended to commercial areas, such as through-the-wall detection [6-9], indoor localization [10-13], biomedical applications [14, 15], and driver assistance [16, 17]. Radars are the key technology in the safety system of a modern vehicle. Automotive radars are the critical sensors in advanced driver-assistance systems (ADAS), which are used for adaptive cruise control, collision avoidance, blind spot detection, lane change assistance, and parking assistance. Evolving from the driver assistance system, a fully autonomous car has much higher demands for its sensors, especially for automotive radars. The topics of this book mainly focus on modern radar technology and its applications on automotive vehicles. The fundamentals of modern automotive radars, including radar system architecture, radar signal processing, noise modeling, and basic detection theorems, are introduced at the beginning by using the frequency-modulated continuous-wave (FMCW) waveform as the example. Due to the scope of this book, details of some theoretical analysis are not discussed in this book. Interested readers may refer to Reference [18] for more theoretical contents. In addition to the radar fundamentals, this book also includes the contents of advanced technologies that have been used in modern automotive radar systems, such as multiple-input and multiple-output (MIMO), sensor fusion, target classification, as well as interference detection and mitigation. These technologies are the keys for supporting ADAS and future fully autonomous vehicles.
In this book, we also recognize that the future of automotive radars should not only address conventional exterior applications but also play important roles for interior applications, such as gesture sensing for human-vehicle interaction, driver/passenger vital signs and presence monitoring, etc. The readers could refer to Reference [19] for additional contents on short-range micromotion sensing with radar technology.
-
Back Matter
- + Show details - Hide details
-
p.
(1)