A Novel Solution to Keep Social Distancing: 3D LiDAR

Quanergy, an optoelectronics company specializing in in 3D detection with lidar, has developed a system that monitors the flow of people in a space to prevent overcrowding and enforce occupancy limits. The system can even track individuals with high body temperatures. The company’s Qortex Flow Management platform won the Pandemic Tech Innovation Award 2020. Following the global wave of Covid-19, Quanergy adapted its platform to provide reliable solutions for social distancing, which remains the most effective strategy for controlling the spread of the virus. Lidar is a time of flight (ToF) sensing technology that pulses a laser at low power and measures the time it takes to complete a round trip between the sensor and a target. “By measuring the time between transmission and reception of emitted light you can measure the distance of an object in real-time,” said Enzo Signore, chief marketing officer at Quanergy. When the lidar sensor emits lights both vertically and horizontally, the collected data is used to generate a 3D image, providing both spatial position and depth information to identify, classify, and track moving objects. “Quanergy’s strategy was originally focused on providing lidar-based solutions for autonomous vehicles. By working with many customers, we’ve seen that the lidar technology was being absorbed by other markets at a much faster rate, not just by the automotive segment. In particular by the 3D drone-based mapping business, by smart cities, smart spaces, security and by the industrial market. And so, over the last couple of years, we have expanded our product strategy to develop products and solutions for these other markets. And now, we’re seeing the benefits of this strategy with accelerating market adoption. Quanergy is in a very good position now to provide not only the hardware sensor, but also the software platform to enable a number of sophisticated applications that the market can rapidly adopt,” said Signore. In terms of security, lidar technology provides a significant reduction in false alarms, a more than five-fold reduction in cost, per coverage area, and significantly higher accuracy than camera-based systems. Enhanced operational intelligence allows you to manage social distancing in terms of space occupation and flow regulation. LiDAR technology The principle behind lidar is basic, but building a lidar systems is not necessarily simple. The formula that engineers use to calculate distance is: the distance of the object = (speed of light x time of flight)/ 2. Lidar measures the flight time used to calculate the distance traveled. The distance traveled is then converted into elevation. These measurements are made by using the key components of an aircraft lidar system, including a GPS that identifies the X,Y,Z position of the light energy and an internal unit of measurement (IMU) that provides the orientation of the aircraft in the sky. One of the complicating factors is that the emitted light can be reflected in different directions. Looking at a tree, some photons will be reflected from  branches, but others will continue through and reflect from the ground — it is possible, even likely, that multiple reflections may occur from one pulse of light. A waveform or light energy distribution is what the lidar sensor measures. Essentially it is very similar to radar technology. Both use parameters like speed, angle, and range. Radar, of course, uses radio waves instead of light. The lidar can provide a 3D image, and thus create the surrounding world in a complete way. Unlike cameras, lidar does not provide any privacy risks. “There are no privacy information violations because there’s no facial recognition,” said Signore. The hardware of a lidar system essentially consists of a transmitter and receiver, and a point cloud that identifies a set of data from the scanned environment, as well as fixed objects such as buildings, trees, and other permanent structures. A data set that can be transformed by a software system to create lidar-based 3D images of a given area. An identifying parameter is the field of view defined as the angle in degrees covered by a sensor. Typically the performance of the lidar sensor is measured in the horizontal and vertical field of view. Another factor to consider in the lidar is the reflectivity of the target. Lidar technology reflects light, some targets reflect better than others and are easier to detect. A white surface reflects more light than a black surface. The latter is therefore very difficult to detect.  Mirrors reflect in a particular direction that may not be the direction of the sensor. Unlike other objects that disperse incident light in various directions. Targets such as traffic signs and license plates reflect a high percentage of light to the receiver and are good targets for lidar sensors. Because of these differences, the actual performance and maximum effective range of a LiDAR sensor may vary depending on the surface reflectance of the target (figure 1).

“There are many unique value propositions we bring to the market. The first one is about the accuracy and the richness of the point clouds. It’s all about data. And so one of the key characteristics that Quanergy provides better than most other solutions is the fact that we can generate a very rich 3D point cloud. For instance, with our products, we can collect up to 1.3 million pulses per second. And that creates a very rich 3D imaging of the world around us,” said Signore. He continued, “The second key differentiator is in the perception software. Having a point cloud is important. But what’s even more important is how to interpret what the sensors are looking at and make sense of it. Therefore, with the perception software that we call Qortex, we can now interpret the objects around the sensor. For instance, we can see people, vehicles, the background, and then we can calculate how many people are in the field of view, not only the XYZ position but also the speed and the direction at which they are moving. The sensor can determine the distance between you and me and tell us if we are at one meter, one and a half, two meters away. So we can use this information also for social distancing applications.  We can provide between 95 to 98% accuracy depending on the type of application, and that’s irrespective of lighting conditions, it can be in a perfect, pitch-black night, or it could be in a full sunny day. Another fundamental difference comes in the overall cost of the solution. In our case, for instance, with our M-Series, we can see and classify objects with over 95% accuracy as far as 70 meters out. Therefore, with one sensor you can cover up to 15,000 square meters. You would need over 100 cameras to cover the same area.” said Signore. Quanergy offers hardware sensors such as M and S Series. M-Series lidar sensors are used for mapping, security, smart city and smart space applications. The S-Series is a true 100% solid-state lidar based on optical phased array (OPA) technology that offers vibration immunity and more than 100,000 hours of the mean time between failures (MTBF). In addition, the economical and scalable CMOS silicon process enables extremely cost-effective mass production. Qortex  platform Quanergy’s Qortex Flow Management platform with LiDAR sensors and AI (Artificial Intelligence) technology provides real-time analysis for people counting, object detection, classification, tracking, monitoring, social distancing, and more. The platform integrates with thermal cameras to identify and track people with high body temperatures. The protection of privacy is guaranteed as no registration is required, and there is no risk of disclosure of personal identifiable information. This solution can be applied for the monitoring of social distancing in companies, retail outlets, airports, factories, distribution centers, and public transport. Each application field has its own complexity and technological challenge. You need to make sure you have enough sensors to cover the entire field of view, but also to bypass occlusions such as columns or other complex environmental, architectural elements that block the field of view (figure 2).

The patent-pending Qortex People Counter platform is based on the Qortex perception software and the S3-2 solid-state lidar sensor. This integrated software/hardware solution incorporates machine learning and 3D perception algorithms to scan the sensor’s field of view, analyze the lidar S3-2 point cloud, and provide anonymous data on people detected in real-time. The S3-2 Series lidar sensor uses SensorFusion technology to enable accurate counting. Qortex DTC (Detect, Track, Classify), the Quanergy Platform brain, enables real-time monitoring, generating a wealth of data including position, direction, speed, and type of detected objects. Through its API, end-users, system integrators and application developers can build a powerful analysis and business intelligence tool to monitor and automate processes. Lidar technology can be a way to impose social distancing. When the distance between individuals is less than recommended, or the number of people in an area exceeds a certain limit, a warning is issued, and staff can be sent. The solution can also be integrated with thermal cameras to identify body temperatures, all while maintaining privacy. The solution can also be used in stadiums, as reopening sporting events to the public could be positive to improve people’s mood in such a stressful period.