LiDAR Object Detection

LiDAR Detection of Real-World Objects

LiDAR Real-World Object Detection

For autonomous vehicles to truly thrive as the next generation of transportation, their ability to detect and differentiate real-world objects must be beyond reproach. Innoviz has developed a solid-state LiDAR system that couples next-generation sensors, advanced algorithms, and state-of-the-art perception software to create the smartest, most aware self-driving vehicles available today. LiDAR sensors use lasers to detect, measure, and generate three-dimensional maps of the surrounding environment. LiDAR technology from Innoviz not only senses the real world, but has the ability to differentiate objects in real-time to create 3D maps with pinpoint accuracy.

How Does LiDAR Detect Real-World Objects?

LiDAR is an acronym that stands for Light Detection and Ranging. Since its inception, LiDAR has historically been used for a wide range of applications that include geological mapping, archaeology, and agriculture. Advanced LiDAR systems, like those developed by Innoviz Technologies, are graduating into uses that can truly hit home.

LiDAR uses infrared laser pulses to measure distances between a centralized point and objects in three-dimensional space. When pulses reflect back to the sensor, they create what is known as a “point cloud”.

Algorithms work to measure the elapsed time for distributed light to return from multiple surface points. LiDAR systems are then able to generate a 3D map based on the reflected points, or the “point cloud”.

How is LiDAR Detection from Innoviz Different?

Innoviz has developed an advanced LiDAR technology that creates realistic 3D environment maps in real time. Both the InnovizOne and InnovizTwo have four individually configurable regions of interest that enables dynamic focus in a limited field of vision to enhance visibility and range without impact to bandwidth, resolution, or frame rate.

Innoviz’s LiDAR also returns multiple reflections per pixel to store multiple points in a 3D environment to deliver a more accurate picture of the surroundings and differentiate objects like raindrops or snowflakes. Contiguous pixels ensure there are no gaps in the scanning pattern to ensure that the system does not miss objects that could fall between point cloud gaps.

Additionally, Innoviz provides an equally distributed field of vision both vertically and horizontally so that no data is lost toward the edges of the field. InnovizOne and InnovizTwo operate exceptionally in both ambient sunlight and poor weather conditions such as rain.

Innoviz Advanced Real-World Object Detection

LiDAR from Innoviz is the most advanced automotive Light Detection and Range system available today, and the only one trusted by BMW for their autonomous vehicles. Our systems are designed to improve users’ safety, deliver better range and differentiation, while fully mapping 3D objects in the real world.

Real-World Detection

Improved User Safety

By advancing how LiDAR systems are able to detect and map three-dimensional space as accurately as possible, Innoviz is improving the safety of autonomous vehicles as they move ever closer to reality. By using advanced technology to create real-time maps, Innoviz is enabling LiDAR to detect and differentiate objects, function well in inclement weather, and foresee potential issues before they arise, resulting in a higher safety standard for users.

Improved Range and Differentiation

Innoviz’s solid-state LiDAR is capable of detection up to two hundred and fifty meters, even in inclement weather. The higher range provides a wider area of focus to detect objects at longer distances in order to anticipate potential risks and allow the system to navigate the vehicle safely. The system is also capable of differentiating objects, identifying different vehicles and environmental hazards, with the ability to map and track unique objects in real-time.

Full 3D Object Maps

With a system that delivers zero gaps between pixels, Innoviz’s LiDAR produces full three-dimensional maps in real-time. The scanning pattern features contiguous pixels that are the foundation of object detection and collision avoidance, creating a complete view of the surrounding environment to ensure that no object is missed and autonomous vehicles are as safe as possible.

View Next