Dimitrakopoulos G. The Future of Intelligent Transport Systems 2020

Sensing and perception systems for ITS Chapter | 1 9

will tackle a promising new field of research by proposing and developing in-
novative approaches to enable a low-cost, reliable, and automotive qualified
LiDAR sensor system. Therefore future advances will generate novel and out-
standing technological know-how. As a further positive effect, the foreseen fun-
damental research and development activities will improve Europe’s reputation
as one of the world leaders in automotive and sensor technological innovation.
Some key figures of state of the art laser scanners are provided in Table 1.6.
Due to the required fine-grained angle and range resolutions, one of the main
problems with LIDAR is that they generate a huge amount of data, which needs
to be locally compressed and prefused before transmitting them to the main unit
for the next level of data fusion with other sensor information. One possible
approach is to provide vector object lists at the interface to reduce the high data
rates, as opposed to sending raw data.

1.4 RADAR

State of the art RADAR sensors either provide several fixed aligned, partly over-
lapping beam lopes or rigid phased-array structures at the receiving frontend.
Typical single RADAR modules from Bosch are shown in Fig. 1.3 (long-range
radar and mid-range radar), which are quite bulky.
Particularly for driving in urban environments, smart beamforming with
high directionality will be necessary to properly capture the motion of other
road users and outside traffic participants in the vicinity. For that reason, solu-
tions for electronic beamforming must be developed to capture the whole ve-
hicle environment and allow surround-vision based on radar (Fig. 1.4).
Beyond advances in hardware and beamforming, the state-of-the-art pro-
grammable radar signal processing will be improved. ADAS system sensors
require a latency below 500 ms. The current baseline measure for interference

TABLE 1.6 State of the art LIDAR specifications.

Sensor

Dimensional

resolution Range

Azimuth

angle

(degree) Accuracy Cycle

Quanergy

M8-

3D 150 m 360 0.05 m, -, 0.03° 33 ms

Ibeo LUX 2D 200 m 110 0.1 m, -, 0.125° 20 ms

Continental

SRL

2D 10 m 27 0.1 m, 0.5 m/s, - 10 ms

Velodyne

HDL-64E S

3D 120 m 360 0.02 m, -, 0.09° 50 ms

Source: de Ponte Müller, 2017