KETI integrates light into chip form, enabling easy channel expansion without optical component modification

For autonomous vehicles to become a reality, multiple technologies are required. Dozens of systems must be fulfilled, including HDA (which automatically maintains inter-vehicle distance), Lane Departure Warning System (LDWS), Lane Keeping Assist System (LKAS), Blind Spot Detection (BSD), AEB, ACC, and others.

The ultimate goal of all these technologies is defined as 'safety'. To realize this technology, development of LiDAR performance and technical capabilities is necessary. LiDAR is a sensor that uses light to determine distance and velocity. Interest in LiDAR is increasing due to growing demand in industrial fields requiring distance information, such as unmanned mobile objects, autonomous vehicles, safe operation, drones, robots, and machine vision.

LiDAR form factors are evolving into fixed-type, beam-splitting, 360-degree scanning, and multi-channel types. The Korea Electronics Technology Institute (hereinafter KETI), a government-funded research institution, is developing all four types and leading efforts to advance Korea's LiDAR market.

To achieve the technological capabilities mentioned earlier, inter-vehicle distance must be identified and controlled. Radar, ultrasonic, and LiDAR are available for this purpose, with radar being most widely used. However, due to radar's several limitations, transition to LiDAR is expected.

Radar is vulnerable at close range when using a single channel, but adding LiDAR improves close-range performance. Additionally, while radar has weak pedestrian recognition capability, LiDAR can recognize pedestrians. However, KETI projects that each technology has its own advantages and disadvantages, and in the future, radar, laser, and camera-based vision will complement each other's functions and be used together.

LiDAR, like radar, uses TOF (Time of Flight) method to determine distance and velocity. KETI developed a 'beam-splitting optical device' where a single laser passes through one lens and splits into three beams. It consists of a 'transmitting lens' that sends light with three beams and a 'receiving lens' that receives light from the left, right, and center.

KETI developed lenses, signal processing units, distance extraction algorithms, relative velocity extraction algorithms, and noise removal algorithms. Information obtained through these is available for real-time confirmation in the UI.

Additionally, KETI developed a compact Laser Range Finder (LRF) small enough to fit in a CCTV dome for supply to major corporations. The system features a modularized 'sensor unit' enabling channel expansion without significant modifications, a 'signal unit' responsible for angular resolution relative to scan speed during rotation, and a 'signal processing unit' that determines signal processing channels, precision levels, and noise/disturbance removal and amplification methods. These three modules can be customized and utilized according to customer requirements for enhanced functionality.

Another advantage is the ease of channel expansion. If operating 16 channels with a 5mm-wide laser, the width would be approximately 8cm. An 8cm transmitting lens would be around 20cm. In other words, significant modification to the sensor unit's optical structure would be necessary. KETI integrated the laser in chip form at 200 micrometers in size. This means channel expansion results in minimal size change, enabling independent sensor expansion. Consequently, signal processing unit size modification is also facilitated. Additionally, using invisible infrared laser eliminates performance degradation in poor visibility conditions such as nighttime or yellow dust.

Lee Han-young, researcher at the Korea Electronics Technology Institute, stated, "Demonstration of partially scannable laser fence and fixed-type LiDAR is possible, and 360-degree single-channel and 360-degree multi-channel development is expected to be completed in the near future."