AI applications and operations
Perception, decisions, action planning and business systems validated within each application.
PHYSICAL AI / CONNECTIVITY
Connectivity forPhysical AI in the field.Connectivity forPhysical AIin the field.
When robots and mobile systems perceive, decide and act in the physical world, they need links between field data and the systems around them. IDY engineers and validates the communications layer through 5G, 60 GHz wireless, Multi-RAT and edge gateways.
01 / IDY'S ROLE
Engineer the linksaround physical systems.Engineer the linksaround physicalsystems.
This page describes IDY’s communications scope. It does not present IDY as the developer of an AI model or robot. The evidence concerns transmission of video and sensor data, links to field edge equipment, and the design of communication paths and recovery.
IDY connectivity and edge links
5G and LTE, 60 GHz wireless, Multi-RAT, gateways, RF and antenna evaluation, and communications path design.
Equipment in the field
Robots, AGVs and AMRs, cameras, vehicles, sensors and controllers.
02 / MEASURED EVIDENCE
Measured performance,with its test conditions.Measured performance,with its testconditions.
Each measurement belongs to its own test environment and path. The figures are not combined into a single end-to-end claim, nor do they measure autonomous robot operation or safety control.
Measured iR800B and 60 GHz link
MAC DL 2.7–3.0 Gbps / ping at or below 1 msMeasured in a lab test and demonstration configuration combining iR800B with 60 GHz wireless. This evaluates a short-range, high-capacity link for workloads such as video. It is not latency across an entire 5G path or robot control system.
View the test configurationLatency measured by path
5G segment 7.5–7.7 ms / E2E 9.3–9.6 msAverage RTT in an n79 100 MHz local 5G test configuration including an L2 overlay. A separate load test also observed higher latency and loss near the configured capacity limit. Capacity and latency must be evaluated together for each use case.
Explore the latency evaluationThermal behavior under load
Approximately 70 min / effective DL around 1.7 GbpsA separate continuous-traffic test recorded modem and PA temperatures. After rising, both broadly settled in the 60°C range. The chart reconstructs approximate temperature trends and does not plot throughput over time.
View the continuous traffic testThe 60 GHz, local 5G and continuous-traffic measurements are separate tests. Figures describe their respective configurations; they are not guaranteed product ratings, robot action latency or performance of a combined end-to-end path.
03 / CONNECTIVITY DESIGN
Design for the siteand its failure modes.Design for the siteand its failuremodes.
Physical AI calls for more than video and sensor bandwidth. We also design for latency variation, mobility and recovery after a link failure. We measure the requirements of each field path before selecting access technologies and devices. AI decision processing and machine safety control require validation within the complete application system.
Measure the field
Assess line of sight, blockage, reflections and mobility paths. Assign short-range, high-capacity 60 GHz and wider-area 5G or LTE their respective roles.
Define traffic paths
Measure bandwidth for video and sensor data, and latency and jitter for monitoring and control paths.
Design for a fault
Evaluate multiple links and device configurations, then validate switching time and recovery behavior under operating conditions.
04 / FIELD REQUIREMENTS
Start with the devicesand data in your field.Start with the devicesand data inyour field.
Share the communication paths and response requirements of your robots, AGVs and AMRs, machine vision or vehicle systems. We can begin with requirements and a proof-of-concept configuration.