IEEE Robotics & Automation Magazine - September 2023 - 29

result, they may require frequent calibration and, subsequently,
possess limited durability. To achieve pure solid-state lidar technologies,
mechanical components, including MEMS, must be
excluded. With this aim, optical phased array (OPA) and flash
lidar sensors have been designed using optical phase modulators
and diffusers, respectively [6]. In particular, in flash lidar sensors,
an instantaneous, uniform flash of light illuminates a wide FoV,
following which, the array of photodetectors receives the returned
signal. They achieve highly efficient, real-time processing due to
synchronized and rapidly updated measurements of distances to
objects, which requires much less data than other mechanical and
quasi-solid-state lidars. Such practical solid-state lidar sensors
have no moving parts due to electronic beam steering technologies
that lead to high durability to shock and vibration, a compact
design, fast scan rate of 100 kHz or more, and low cost. Furthermore,
such advantages make solid-state lidar sensors inherently
more cost-efficient to produce, enabling mass production and
commercial viability. However, there are also some disadvantages,
such as a relatively narrow FoV and fragility to external light.
RADAR
Instead of light waves with wavelengths of 600-1,000 nm used
by lidar, radio detection and ranging (radar) sensors employ
radio waves to detect objects. Radio waves are a type of electromagnetic
radiation with wavelengths longer than infrared light,
ranging from 1 mm (300 GHz) to 10,000 km (30 Hz). The
transmitter of a radar sensor emits a radio wave, which is
reflected off an object and returns to the receiver with information
about the position and velocity of the object. Radar sensors
are gradually being integrated into applications across various
fields and are available in designs of varying specifications,
sizes, and prices. Additionally, radar sensors primarily function
based on the properties of continuous and pulsed waves. To
appropriately represent continuous and pulsed-wave radar sensors
in this study, a frequency-modulated continuous wave
(FMCW) radar sensor and an impulse radio ultrawideband
(IR-UWB) radar sensor were considered in view of their widespread
use in robotics and industrial applications [7], [8].
CW radars measure range by transmitting and receiving
radio waves simultaneously with separate antennas. In particular,
FMCW radars use FMCWs to improve their measurement accuracy
and discriminate among multiple targets. Target velocity can
also be detected using frequency modulation. In contrast, pulsedwave
radars emit high-power short pulses and receive echo signals
reflected by targets to measure range. They interleave transmission
and reception signals with their single antenna. When the
two sensors are compared for the same peak amplitude, a CW
radar sensor has a better signal-to-noise ratio because it produces
a higher signal power by emitting signals for a longer duration.
As the radar is based on electromagnetic radiation of a longer
wavelength than that of the lidar, the former is obviously more
effective in applications where the detection distance is more
important than the precise image of an object. Furthermore, radar
experiences relatively small propagation attenuation; hence, it can
properly operate even in environments of snow, clouds, or fog [9].
Unlike lidar, radar can see through materials and obtain the necessary
data without a line of sight [10]. Recently, radar sensors have
also been employed to perform human detection without a serious
invasion of personal privacy, considering that precise images of
objects are not provided. However, radar sensors have a limitation
in that they lack precision when detecting small-sized objects,
resulting in low resolution. This limitation can be overcome by the
complementary usage of lidar sensors, which is discussed in detail
in the " Overall Comparison and Complementary Strategies " section.
The schematics of each sensor are presented in Figure 1.
PERFORMANCE COMPARISON
Three comparative experiments were performed to examine
object detection ability, mapping performance, and environmental
robustness. An object detection test was conducted to quantitatively
evaluate the extent of influence on the sensors by the type
or size of object material. This evaluation is significantly useful
SEPTEMBER 2023 IEEE ROBOTICS & AUTOMATION MAGAZINE
29
IMAGE LICENSED BY INGRAM PUBLISHING
IEEE Robotics & Automation Magazine - September 2023

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - September 2023
