In this presentation, we will provide an overview of the findings of the products that use LiDAR data for real-time object identification, classification and tracking in roadside infrastructure, with a focus on detecting pedestrians. As a metrology institution, the National Physical Laboratory (NPL) will use a measurement perspective to investigate the product. The perception boxes are a black box to NPL, there is no access to the algorithms, however, we can still access the performance difference between algorithms. You can learn from our findings that a few error factors could contribute to the incorrect output. For a self-driving vehicle, receiving a false alarm or ignoring a passed-out body on the floor causes significant issues to credibility. These measurements could also be input for those users who intend to use the perception boxes as part of their infrastructure and the next iteration of the perception boxes.