Powering the Physical AI Revolution by Transforming Motion into Intelligence<\/b><\/p>\n<\/div><\/div><\/div><\/div><\/div> While the world focused on Generative Al (text\/images), a “quiet revolution” has begun: Physical Al.<\/span><\/p>\n Definition: <\/b>Physical AI enables autonomous systems such as robots, self-driving cars, and smart spaces to perceive, understand, and act in the real world.<\/span><\/p>\n \u201cLarge language models are one-dimensional… Image- and video-generation models are two-dimensional… None of these models can understand or interpret the 3D world.\u201d <\/b>\u2014 Source: CSET Physical AI Report<\/a><\/i><\/span><\/span><\/p>\n The next phase of AI is “Physical AI”-autonomous systems interacting with the real world. Since motion in the world happens in 3D, current 2D sensing technologies (such as cameras and radars) cannot support this future due to a lack of depth perception.<\/span><\/i><\/span><\/span><\/p>\n<\/div><\/div><\/div><\/div><\/div> 3D LiDAR provides depth, dimension, position, and speed, thus enabling the measurement of motion in real-time. LiDAR, therefore, becomes the key enabler for Physical AI, expandable across a \u201cUnified Citywide Data Space\u201d that enhances safety and creates monetizable data assets.<\/span><\/p>\n<\/div><\/div><\/div><\/div><\/div> \n The Motion Problem: A Universal Challenge<\/span><\/strong><\/p>\n From protecting data centers to managing airport congestion or guiding autonomous vehicles, the core issue is the same: <\/span>motion is difficult to track, interpret, and respond to, especially when traditional technologies are used.<\/strong><\/p>\n Physical Security – When Every Swaying Tree Looks Like a Threat<\/strong><\/p>\n Perimeter security is a high-stakes game. Whether it\u2019s a <\/span>data center<\/a><\/u>, utility substation,<\/span><\/strong> or <\/span>rail corridor,<\/strong> intrusions can cost millions and threaten business continuity.<\/span><\/p>\n Despite expansive investments in video analytics, <\/span>cameras don\u2019t know what matters.<\/strong><\/p>\n They react to any motion, whether it\u2019s a <\/span>dog wandering past, a gust of wind rustling tree branches,<\/strong> or <\/span>cars driving by.<\/strong> This results in:<\/span><\/p>\n The fundamental problem is the passive 2D technology used, which is not designed to accurately capture motion in a 3D world, making it difficult to precisely locate and track intruders.<\/p>\n To track a moving person across a site, operators must stitch together multiple screens, with shadows, poor lighting and occlusions, trying to anticipate where someone might go next, often too late to act.<\/p>\n Radar has a different problem.<\/a><\/strong> It\u2019s typically optimized for <\/span>either low-speed motion (people)<\/strong> or <\/span>high-speed motion (vehicles)<\/strong>, but not both. In environments like utilities, this becomes a dangerous limitation.<\/span><\/p>\n These limitations not only impact the security of buildings and the safety of people, but<\/span> also contribute to escalating theft and pilferage. <\/strong>According to <\/span>Wired<\/a>, \u201cMetal Prices Are Soaring. So Is Metal Theft.\u201d Metal theft is labeled a \u201cglobal epidemic,\u201d with copper repeatedly stolen, even from rail infrastructure, because legacy sensors just don\u2019t catch thieves moving among vehicles and trains.<\/span><\/p>\n Smart Spaces – Airports Are Rivers of People. Cameras Can\u2019t See the Flow.<\/strong><\/p>\n Aeropuertos<\/a> are among the most complex motion environments in the world. At any given time, <\/span>thousands of passengers<\/strong> are walking, pausing, turning around, entering queues, or clustering near gates and around columns, counters and other occlusions.<\/span><\/p>\n And airports operators need to precisely answer key questions:<\/p>\n Cameras can capture images but cannot effectively track the motion of passengers across large terminals. Due to their limited coverage and lack of meshing capabilities, hundreds if not thousands of sensors would be needed, making the cost of the required networking, cabling and installation infrastructure unaffordable.<\/p>\n Cameras\u2019 lack of scalable intelligence impacts Airports\u2019 ability to:<\/p>\n And because they collect identifiable imagery, cameras introduce <\/span>privacy and regulatory challenges<\/strong>, especially in international terminals. This often results in <\/span>limited deployment<\/strong> and <\/span>underutilized data.<\/strong><\/p>\n Gartner predicts that<\/span> \u201c80% of 3D imaging sensors for smart spaces will use LiDAR or mm Wave radar as the primary vision imaging sensor by 2030.\u201d<\/strong><\/em><\/p>\n Transportation – Motion Overload<\/strong><\/p>\n In rail and metro stations, motion sensing becomes even more difficult. Imagine trying to answer a seemingly simple question: <\/span>\u201cHow many people just boarded that train?\u201d<\/strong><\/p>\n Now add these layers of complexity:<\/p>\n Cameras struggle with this environment. Reflections from train windows distort images. Occlusions make it hard to track individuals. Lighting conditions vary wildly between platforms and tunnels. And again, privacy concerns limit where and how cameras can be used.<\/p>\n![](\"https:\/\/quanergy.com\/wp-content\/uploads\/iStock-1573782786-1200x495.jpg\" "\\"Motion")
<\/span><\/div><\/div><\/div><\/div><\/div>\n
\n
\n
\n
![](\"https:\/\/quanergy.com\/wp-content\/uploads\/Market-Breakdown-800x800.png\" "\\"Market")
<\/span><\/div><\/div><\/div><\/div><\/div>![](\"https:\/\/quanergy.com\/wp-content\/uploads\/Physical-Security-SAM-800x700.png\" "\\"Physical")
<\/span><\/div><\/div><\/div><\/div><\/div>