LiDAR

What is LiDAR?

LiDAR (Light Detection and Ranging) is an optical measurement technique that determines distance by emitting laser pulses and timing the reflected light, allowing a sensor to build a 2D outline or 3D point cloud of its surroundings. In industrial automation, LiDAR acts as the eyes of AGVs, autonomous mobile robots, cobots and safety laser scanners. Within OT environments it delivers real-time spatial information that lets systems navigate, detect obstacles and protect people, which makes it a building block of Industry 4.0, Industrial AI and Cyber-Physical Systems.

🧠 How does LiDAR work?

LiDAR works much like radar, but uses light instead of radio waves. The system emits thousands to millions of laser pulses per second and measures how long each pulse takes to return from an object. The distance follows from that time; by rotating or sweeping the beam, the sensor builds a picture of its entire surroundings.

The basic principle in five steps:

1. The laser transmitter emits a short pulse of light.
2. The pulse hits an object and is diffusely reflected.
3. A photodetector captures a small fraction of the light.
4. The electronics measure the time difference between emission and reception.
5. The embedded controller calculates the distance and combines all measurements into a scan.

A LiDAR system typically consists of:

📏 What is the time-of-flight principle?

Most industrial LiDAR systems measure according to the time-of-flight (ToF) principle. The distance follows from:

d = (c × t) / 2

Dividing by two corrects for the outward and return path. A worked example shows how precise the electronics must be: an object at 10 metres gives a round-trip time of about 66.7 nanoseconds, and every nanosecond of timing error means roughly 15 centimetres of distance error. LiDAR sensors therefore use very fast timing circuits and often average several pulses.

Besides pulsed ToF there is phase-shift measurement (a continuously modulated beam, for short ranges with high precision) and FMCW (frequency-modulated continuous wave), which also measures an object’s velocity directly through the Doppler effect and is less sensitive to interference from other sensors.

🏗️ What types of LiDAR exist?

Solid-state LiDAR is gaining ground thanks to less wear, a more compact form factor and usually lower power consumption. Mechanical 3D scanners still offer the widest field of view.

When selecting a sensor, these are the performance characteristics to compare:

The wavelength matters a great deal. Systems at 905 nm use inexpensive silicon detectors, but this light reaches the retina, which limits the permitted power. Light at 1550 nm is largely absorbed by the cornea and lens, allowing higher power and longer range, at higher cost and with more degradation in rain and fog. Industrial LiDAR sensors are almost always designed as laser class 1 under IEC 60825-1, meaning they are eye-safe under normal use.

🛡️ What is the difference between navigation LiDAR and a safety laser scanner?

This is the most important distinction in practice. An ordinary LiDAR for navigation or mapping is a measuring instrument. A safety laser scanner is a safety component designed under part 3 of IEC 61496 as an AOPDDR (active opto-electronic protective device responsive to diffuse reflection) and certified for use in a safety function. Only the latter may be used to protect people.

A safety laser scanner monitors configurable protective fields and warning fields. When someone enters the warning field, the machine or vehicle slows down; when the protective field is interrupted, a safe stop follows. Vehicles switch dynamically between field sets depending on speed and direction of travel. The outputs go to a safety PLC or a safety relay, alongside emergency stops and light curtains. The whole falls under functional safety according to ISO 13849 or IEC 62061.

🔧 How do you deploy a safety laser scanner in practice?

A typical approach for a machine or an AGV project:

1. Risk assessment - use a risk assessment to determine the hazards and the Performance Level the safety function requires.
2. Calculate the safety distance - ISO 13855 gives the formula S = K × T + C. For an approaching person K = 1600 mm/s; T is the total stopping time of scanner, control system and machine; C is an allowance that, for a horizontal scan field, depends on the mounting height.
3. Configure the fields - draw protective fields with margins for measurement tolerance and reflective surfaces; for vehicles, one field set per speed.
4. Integrate - connect via OSSD or a safety bus to the control system and record the configuration with its checksum.
5. Validate - test every field with a test object, document the results and repeat after any change.
6. Maintain - clean the front window, check the mounting and schedule periodic calibration and function tests.

For AGVs and AMRs, ISO 3691-4:2023, the standard for driverless industrial trucks, applies as well. It requires, among other things, person detection in zones where people are present and braking performance that brings the vehicle to a halt within the detection range. In the Netherlands and the wider EU these systems currently fall under the Machinery Directive; from 20 January 2027 the Machinery Regulation (EU) 2023/1230 applies, which also sets requirements for protection against corruption of control systems and software.

🏭 Where is LiDAR used in industry?

SLAM (Simultaneous Localization and Mapping) is the technique in which a vehicle builds a real-time map of its surroundings while determining its own position within it. In robotics LiDAR is popular for this because it measures distance directly and does not depend on ambient light. Modern AMR platforms combine LiDAR with vision, ultrasonic sensors, AI and machine learning into a sensor-fusion platform. 3D scans and periodic measurements also support inspection and predictive maintenance, for example by tracking deformation of racking or wear on conveyors.

📡 How do you integrate LiDAR into the OT network?

Navigation LiDAR usually delivers a continuous data stream over Industrial Ethernet, often using the manufacturer’s own UDP or TCP protocols. Status and measured values reach the PLC via ProfiNET, Ethernet IP or Modbus TCP, and further up SCADA, MES or a fleet manager via OPC UA or MQTT. Raw point clouds are too large for that route; they are processed locally on edge systems, embedded controllers, industrial IPCs or GPU systems.

A multi-channel 3D LiDAR easily produces tens to hundreds of megabits per second. Real-time processing therefore demands low latency, limited jitter and sufficient bandwidth, especially when several sensors share one network segment.

🔐 What cybersecurity risks does LiDAR have?

Because LiDAR is part of connected OT systems, it also carries cybersecurity risks. From 2015 onwards, researchers demonstrated that LiDAR can be fooled with inexpensive equipment by injecting fake pulses.

Measures:

• Network Segmentation - place sensors and vehicles in their own zones and conduits in line with IEC 62443.
• Firewall and IDS - filter and monitor traffic to configuration and firmware ports.
• Zero Trust - give only authorised engineering workstations access to the configuration.
• Firmware Signing and patch management - install only signed firmware and keep the sensors in your asset inventory.
• Plausibility checks - compare LiDAR with other sensors so that a single manipulated source never decides alone.

Network-connected sensors placed on the EU market after 11 December 2027 must comply with the Cyber Resilience Act.

⚠️ What are the benefits and limitations?

Benefits: high distance accuracy, works in the dark, direct 3D information, real-time detection and certified variants for machine safety.

❓ Frequently asked questions

Is LiDAR harmful to the eyes?

Industrial LiDAR sensors are almost always designed as laser class 1 under IEC 60825-1. That means LiDAR is eye-safe under normal use, even when looking directly into the beam. Never use optical instruments to look into an opened or faulty sensor.

Can I use an ordinary LiDAR to protect people?

No, protecting people requires a safety laser scanner certified as type 3 under IEC 61496-3. An ordinary navigation LiDAR has no guaranteed response time or fault detection and does not count towards a safety function under ISO 13849.

What is the difference between 2D and 3D LiDAR?

A 2D LiDAR scans a single horizontal plane and is the standard for safety laser scanners and AGV navigation. A 3D LiDAR scans multiple planes or a whole field of view and delivers a point cloud, useful for mapping, obstacles at different heights and volume measurement.

Does LiDAR work in dust, fog or outdoors?

LiDAR is largely independent of ambient light, but dust, fog, steam and rain scatter the laser pulses. Safety laser scanners therefore use particle-filtering algorithms and often an outdoor variant, but in harsh conditions availability can drop because of nuisance stops.

How accurate is LiDAR?

Industrial LiDAR sensors typically measure with an accuracy of a few millimetres to a few centimetres, depending on distance, reflectivity and measuring principle. Phase-shift measurement is more accurate at short range, while pulsed time-of-flight reaches further.

Does LiDAR fall under the Cyber Resilience Act?

A LiDAR sensor with a network interface or software is a product with digital elements and therefore falls under the Cyber Resilience Act. The obligation to report actively exploited vulnerabilities applies from 11 September 2026, the full requirements from 11 December 2027.

📌 In summary

LiDAR measures distance with laser pulses and acts as the eyes of AGVs, AMRs, cobots and safety laser scanners in industry. For protecting people, use only scanners certified under IEC 61496-3, calculate the safety distance with ISO 13855, and secure sensors and their configurations as full OT assets in line with IEC 62443.