Demonstrations

The animation opens on the truck kit's complete sensor architecture, the first system designed to run on the box. This establishes what the hardware covers before any individual component is explained.

Two retrofit mirror towers, one above each wing mirror, each hold a thermal camera, a road-facing camera, and a 120-degree side camera. Together they cover the long flanks of the vehicle that drivers cannot easily see.

Inside the cab, a forward dash cam records events ahead while a driver monitoring camera watches for fatigue. These two sensors address risk that originates with the driver rather than the road.

Front and rear 4D radar give range and velocity at both ends of the truck in rain, dust and glare, the conditions in which cameras alone see least.

The animation sweeps the coverage cones. Where radar and camera fields overlap is where a cyclist or an animal is picked up and held across frames.

The layout makes one simple claim: no quarter of a heavy truck is left unobserved.

The animation runs the same workload twice, side by side, changing only how each core reaches memory. That single difference drives everything that follows.

The current arrangement puts eight processing cores behind a single data engine that fetches one descriptor at a time, so seven cores sit idle while one is fed.

The utilization column on the left never clears 12%, and most cores sit at almost nothing for the entire run, showing how much compute is wasted by the bottleneck.

On the right, the only change is that every core now owns its own load engine and pulls from banked static memory in parallel, with no other modification to the design.

All eight cores climb to around 59% utilization and hold there, reached by nothing more than opening the path to memory.

The frame that costs 470 milliseconds in the current arrangement lands in about 85 milliseconds in the parallel-engine simulation, a direct result of the memory path change alone.