New Possibilities for Autonomous Driving Opened by 200 m (656.2 ft) LiDAR Point Clouds

Table of Contents

• LiDAR point clouds and autonomous driving

• Evolution of LiDAR sensor range

• New possibilities brought by 200 m (656.2 ft) LiDAR

• Challenges of LiDAR deployment

• Conclusion

• Simplified surveying with LRTK

• FAQ

LiDAR point clouds and autonomous driving

LiDAR (Light Detection and Ranging) is a sensor that uses laser light to measure distances to the surroundings. It emits pulsed laser light and calculates distance from the time it takes for the light to hit an object and return. The point cloud data obtained by LiDAR is the collection of countless such distance measurement points. A point cloud represents the environment as a 3D collection of points, detailing the shapes and positions of objects such as buildings, vehicles, and people.

In autonomous vehicles, LiDAR plays the important role of the “eyes.” Vehicle-mounted LiDAR emits hundreds of thousands of laser pulses per second, scanning the surrounding 360 degrees in real time to generate point cloud data. From this point cloud, it can detect precise distances, directions, and sizes of other vehicles, pedestrians, and obstacles. LiDAR point clouds are also used for self-localization by matching them against high-precision 3D maps (knowing the vehicle’s exact current position). Where cameras and satellite positioning (GPS) alone struggle to achieve centimeter-level self-positioning, the detailed 3D data from LiDAR can make that possible (centimeter-level self-positioning (cm level accuracy (half-inch accuracy))).

Other sensors such as cameras and millimeter-wave radar are also used in autonomous driving, but each has strengths and weaknesses. Cameras excel at recognizing signal colors and text on signs but cannot directly measure distance and perform poorly at night or in fog/backlighting. Radar can detect distant objects using radio waves and is less affected by weather, but its low resolution makes detecting small obstacles and determining object shapes difficult. LiDAR measures distance via laser reflections, enabling extremely high-precision 3D measurement and actively illuminating the environment even in darkness. Because point clouds capture accurate distances and contours to objects, LiDAR can detect small obstacles that other sensors might miss and thus contribute to safer driving decisions.

In this way, LiDAR complements cameras and radar, giving autonomous vehicles detailed environmental awareness. Therefore, LiDAR installation is emphasized in advanced autonomous vehicles at Level 3 and above, and it is positioned as a “sensor that holds the key to autonomous driving.”

Evolution of LiDAR sensor range

Conventional automotive LiDAR sensors typically detected objects up to around 100 m (328.1 ft). However, in recent years, that measurement range has been rapidly extending. The latest technologies have produced and demonstrated long-range LiDAR capable of detecting objects 200 m (656.2 ft) away. This represents roughly double the range of earlier LiDAR and is an important breakthrough for advanced autonomous driving.

Why was measuring as far as 200 m difficult? One reason is the need to detect extremely weak laser reflections from distant objects amid noise such as sunlight. As distance increases, the returned light from the target becomes extremely weak, requiring highly sensitive, noise-resistant photoreceivers in the sensor. Additionally, vehicle-mounted LiDAR faces limits on laser output for eye safety, making it historically difficult to increase emission power. To overcome these constraints, companies have implemented various innovations. For example, changing the laser wavelength from the commonly used 905 nm to the 1550 nm band allows higher output while maintaining eye safety to achieve longer ranges. Other approaches include advanced ADC (analog-to-digital conversion) circuits that digitally average signals to reduce noise, and hybrid methods that combine near-range and long-range detection modes. As a result, performance is being realized that can detect small, low-reflectivity objects at 200 m (656.2 ft) even under strong sunlight.

The practical deployment of long-range LiDAR directly enhances autonomous vehicle capabilities. Long-range detection on the order of 200 m (656.2 ft) is especially anticipated for highway autonomous driving at high speeds and for Level 4 and above systems that require larger safety margins. Ranges that previously required multiple LiDAR units may now be covered with fewer sensors, enabling broad-area monitoring. Expansion of LiDAR measurement range is a key factor in widening the scope of autonomous driving.

New possibilities brought by 200 m (656.2 ft) LiDAR

The arrival of LiDAR that can detect up to 200 m (656.2 ft) expands the capabilities and use cases of autonomous vehicles. Below are the main benefits enabled by long-range LiDAR.

• Improved safety at high speeds: When traveling at speeds exceeding 100 km/h (about 28 m/s), it is extremely important to perceive the road ahead as far as possible. Detecting obstacles 200 m (656.2 ft) ahead provides nearly twice the reaction time compared to seeing only up to 100 m (328.1 ft). For example, if traveling at 100 km/h (about 28 m/s), detecting a hazard 200 m (656.2 ft) ahead allows deceleration to begin roughly 7 seconds earlier, enabling safe stopping or avoidance without emergency braking. This can reduce accident risk at high speeds that older LiDAR systems could not adequately address.

• Early detection of small obstacles: Small obstacles such as fallen tire fragments or small animals on the road are easily missed by conventional sensors and are dangerous. Long-range LiDAR can capture small objects that return only weak reflections at longer distances, allowing earlier recognition and response. Even dark-colored debris on a nighttime highway can be detected at 200 m (656.2 ft), giving the driver (or vehicle AI) time to decelerate or avoid the object safely and preventing sudden steering maneuvers.

• More time for route planning: With sensing extended farther ahead, the vehicle’s route planning gains more leeway. Early awareness of lane congestion, pedestrians or cyclists on the road, or lane restrictions due to construction allows smoother lane changes or gradual deceleration well in advance. As a result, unnecessary abrupt braking or lane changes decrease, providing a safer and more comfortable ride for occupants.

• High-precision 3D map creation and updates: LiDAR point cloud data is also valuable for creating and updating high-precision 3D maps (HD maps) for autonomous driving. A sensor capable of detailed scanning up to 200 m (656.2 ft) can efficiently map a wide area while driving. Buildings, road signs, and traffic lights several blocks ahead can be captured in a single pass as point clouds, enabling rapid detection of new objects or changes to be reflected in maps. In the future, autonomous vehicles could share their long-range point cloud data to the cloud and share road conditions with other vehicles in real time, enabling digital twin-style applications.

The expanded field of view afforded by 200 m (656.2 ft)-class LiDAR thus holds great potential to dramatically improve the safety and utility of autonomous driving.

Challenges of LiDAR deployment

While long-range LiDAR offers major advantages for autonomous driving, several challenges remain to be solved for practical deployment.

• Cost and integration issues: LiDAR units require advanced optics and electronics, so they remain costly; price reductions are essential for widespread adoption in mass-market vehicles. Rotating LiDARs have historically been large, making integration into vehicle design difficult. However, solid-state LiDAR and other miniaturization and cost-reduction advances have made it possible to place sensors unobtrusively in grilles or on roofs. Further economies of scale from mass production are expected to reduce costs.

• Effects of adverse weather and environment: Because LiDAR uses laser light, its performance is influenced by rain, fog, snow, and other weather conditions. Dense rain or fog particles scatter and attenuate laser beams, reducing detection range and increasing noise points. Lens areas in environments prone to dirt (splash or dust) require regular cleaning. In such situations, combining LiDAR with other sensors like millimeter-wave radar, or using filtering techniques to reduce noise in LiDAR point clouds, are common mitigation strategies.

• Data processing and software: High-resolution LiDAR can output millions of points per second, so real-time processing requires powerful computing performance and sophisticated algorithms. Software for sensor fusion—integrating point clouds with camera images and radar data to correctly interpret the environment—is also critical. Advances in AI and dedicated semiconductor chips have gradually enabled analysis and integration of these large datasets, but further software improvements and validation are required to ensure safe driving.

The industry is addressing these challenges and steadily lowering the barriers. As costs fall and the technology matures, the adoption of long-range LiDAR is likely to accelerate, bringing closer a future where it becomes standard equipment on autonomous vehicles.

Conclusion

The emergence of LiDAR that can perceive surroundings up to 200 m (656.2 ft) is a major driving force opening new possibilities for autonomous driving. Vehicles with a wider field of view can travel more safely on highways, detect small risks they would otherwise miss, and enable smoother, more deliberate driving. While LiDAR alone does not solve every issue and challenges such as sensor fusion and cost reduction remain, its technical value is growing year by year.

The ability to acquire point cloud data at a level once unimaginable—200 m (656.2 ft)—gives autonomous vehicles “eyes that see farther.” This is a capability beyond human drivers and a powerful catalyst for achieving truly reliable autonomous driving. As long-range LiDAR technology advances and becomes more widespread, autonomous driving will attain even higher levels of safety and trustworthiness, transforming our mobility and daily lives.

These high-resolution 3D measurement techniques using LiDAR are also expanding beyond autonomous driving. For example, in infrastructure inspection and civil engineering surveying, LiDAR is being used to streamline historically labor-intensive field measurements. Next, we briefly introduce simplified surveying using the solution called LRTK.

Simplified surveying with LRTK

Finally, as an example outside autonomous driving that leverages LiDAR point clouds, we introduce simplified surveying using LRTK. LRTK combines GNSS RTK positioning technology with a high-resolution 3D laser scanner to provide a solution that anyone can use in the field to obtain high-precision point cloud data. Traditionally, 3D surveying of terrain and structures required placing target markers and skilled technicians, but with LRTK no markers are required, and centimeter-level coordinates can be automatically attached during scanning. No tedious preparatory work is needed: simply power on the equipment and walk while aiming the laser to complete high-precision surveying on site. This greatly reduces preparation time and enables speedy surveys.

LRTK LiDAR also excels at long-range measurement and can produce high-resolution point clouds of structures up to 200 m (656.2 ft) away. It supports surveying wide areas and complex terrain, and can capture thin objects such as power lines comprehensively as point cloud data. No special skills are required: by scanning while watching the device screen, users can collect site data with “zero missed captures.” Because absolute coordinates are added to the point clouds in real time, multiple point clouds measured on different days automatically align spatially, eliminating the effort of merging point clouds in post-processing.

Another feature of LRTK is its ability to survey large areas in a short time. Even several-hectare sites can be point-cloud scanned in about one hour, and it is usable for urban surveys where drones cannot be flown. Data use after measurement is also simple: scanned point clouds can be uploaded to the cloud for centralized management and accessed via browser-based tools to check coordinates, measure distances and slopes, and generate cross-sections. Stakeholders can share data without installing dedicated software, enabling rapid analysis and reporting of survey results.

In this way, simplified surveying with LRTK dramatically improves surveying efficiency and accuracy through LiDAR technology. It is a good example of how advanced sensor technologies developed for autonomous driving are beginning to provide new value in infrastructure management and construction.

FAQ

Q1. What is LiDAR? A1. LiDAR (Light Detection and Ranging) is a sensor that uses laser light to measure distances to targets. Automotive LiDAR can capture the distances and shapes of surrounding objects in 3D and serves as an important “eye” alongside cameras and radar for autonomous vehicles.

Q2. What is point cloud data? A2. Point cloud data is the collection of countless measurement points obtained by LiDAR. Each point has spatial coordinates (X, Y, Z), and because the points look like a cloud, they are called a “point cloud.” By analyzing this 3D point cloud data, a computer can recognize the positions and shapes of surrounding objects.

Q3. What changes if a sensor can detect up to 200 m (656.2 ft)? A3. Extending sensor range improves safety and provides more time for response in autonomous vehicles. For example, detecting hazards 200 m (656.2 ft) ahead while driving at high speed gives about twice as much time for deceleration or avoidance compared to a 100 m (328.1 ft) range. It also enables earlier recognition of distant small obstacles and road conditions, allowing smoother and more predictive driving.

Q4. Does LiDAR work at night or in bad weather? A4. At night, LiDAR can detect objects because it actively emits laser light. However, in adverse weather such as rain, fog, or snow, laser beams are scattered and attenuated, which can reduce detection range and increase noise. Therefore, in poor conditions it is desirable to complement LiDAR with other sensors like radar.

Q5. Is LiDAR necessary for autonomous driving? A5. LiDAR is increasingly being adopted for advanced autonomous driving systems that prioritize safety. Basic driver assistance can be achieved with cameras and radar alone, but LiDAR allows more accurate measurement of distances and 3D shapes, enabling more reliable detection and decision-making. In particular, LiDAR is becoming effectively standard for Level 3 and higher autonomous driving.

Q6. What is LRTK? A6. LRTK is a surveying system that combines GNSS (satellite positioning) real-time kinematic (RTK) technology with high-performance LiDAR. It consists of a dedicated LiDAR device, receivers, and an app, enabling easy, high-precision point cloud surveying without complex setup. For example, using LRTK you can quickly 3D-measure a large site and automatically obtain coordinates for the acquired point cloud, yielding accurate survey results even without specialized knowledge.