Technology

The path to Level 4 in the terminal.

Level 4 means no driver to fall back on. We get there the way the leaders in autonomy do: automotive-grade compute in the vehicle, a digital twin to test against, and a safety architecture that assumes things will fail.

Concept visualisation: the General Axis vehicle carrying a container through a terminal lane at night in heavy rain, headlights on
Around the clock

Night, rain, peak hour.
The same move.

No shift change and no lights-out. Camera, lidar and 77 GHz radar overlap so the vehicle keeps working in the sea fog, spray and darkness ports actually operate in.

01 Level 4

What Level 4 actually means.

Under SAE J3016, a Level 4 system performs the whole driving task and its own fallback within a defined operational design domain. Nobody is expected to take over; if conditions leave that domain, the vehicle brings itself to a safe state.1

SAE levelWho drivesFallbackWhere it operates
Level 2 Driver, supervised assistance Driver Anywhere the feature works
Level 3 System, while engaged Driver on request Limited conditions
Level 4Our target System System reaches a safe state A defined operating domain
Level 5 System System Everywhere

Simplified from SAE J3016_202104. Our operating domain is the mapped, geofenced area of each terminal we deploy in.

02 The domain

Why the terminal is where Level 4 works first.

The industry's Level 4 programmes, from robotaxis to highway trucks, all start by drawing a tight operating domain.2 A container terminal offers one of the tightest there is.

01

A private, mapped site

A terminal is fenced, surveyed and controlled. The operating domain can be drawn precisely and enforced with geofences.

02

Bounded speed and routes

Quay-to-yard moves repeat on known lanes at low speed, far from the open-ended complexity of public roads.

03

Rules that already exist

Terminals segregate people and machines by design, and their operating systems already dispatch every container move.

04

Value in every hour

Ports run 24/7. A vehicle that never changes shift turns autonomy directly into throughput and lower cost per move.

03 Pipeline

Three computers, one data flywheel.

NVIDIA describes physical AI as three computers working in a loop: one to train the models, one to simulate the world they will face, and one to run them in the machine. Every mile and every edge case feeds the next round.3

01 · Train

Data-centre GPUs

Perception and planning networks are trained on real terminal drives and synthetic scenes, then optimised for the vehicle with TensorRT.

02 · Simulate

Omniverse and Cosmos

A digital twin of the terminal tests every software build in closed loop, across conditions too rare or dangerous to stage for real.

03 · Drive

The vehicle computer

The validated software runs on the vehicle. What it encounters is logged and returned to training, closing the loop.

04 Simulation

Test a million moves before the first one.

A Level 4 vehicle has to be right in situations it may meet once a year. Simulation is how those situations get tested thousands of times before deployment.4

Omniverse · OpenUSD

A digital twin of the terminal

Omniverse builds physically accurate digital twins on OpenUSD. Its AV simulation blueprint replays real sensor logs, produces ground truth and runs closed-loop tests, with physically based camera, lidar and radar models.

Cosmos

The scenarios we cannot wait for

Cosmos world foundation models generate and transform driving scenes, turning one recorded move into the same move in fog, rain, glare or darkness, so rare and dangerous cases are trained on before they happen. 5

Isaac Sim

Robots and handoffs in simulation

Isaac Sim, also built on Omniverse, simulates robots and generates synthetic training data, a fit for modelling container handoffs, docking alignment and the machines a vehicle works beside. 6

The simulation pipeline is on our roadmap; see the status table below.

05 Safety architecture

Two layers between a fault and a hazard.

The first is the platform: automotive-grade compute engineered toward functional-safety standards, with redundant power and a safety microcontroller alongside the main processor.

The second is ours: an isolated safety channel, outside the main computer, that checks every motion command against geofences and proximity limits and can stop the vehicle on its own, backed by a hardware emergency stop and contact edges.

Safety and compliance on the platform page
Minimal-risk condition
Controlled stop to a known-safe state on any detected fault
Redundancy
Independent sensing, compute, power and braking paths
Sensor overlap
Camera, lidar and 77 GHz radar cover each other's weaknesses
Operating domain
Geofenced zones with per-zone speed and behaviour limits
Traceability
Full sensor and decision logs for every mission
Standards
Engineered toward ISO 26262, ISO 21448 (SOTIF) and ISO 3691-4
06 Status

Where we are today.

An honest view of what is running, what we are building and what comes next.

ComponentRoleStatus
Perception, planning and control stack Autonomy software In development
Independent safety monitor and e-stop Fail-safe channel In development
Fleet dispatch and monitoring Operations software In development
Omniverse terminal digital twin Closed-loop simulation Roadmap
Cosmos synthetic scenario generation Training data Roadmap
Isaac Sim handoff and docking models Robotics simulation Roadmap
Work with us

Bring a terminal into the loop.

Pilot partners shape the operating domain, the first routes and the digital twin we validate against.