PlusAI has released SuperDrive 6.0, the latest version of its autonomous driving software engineered for commercial-scale deployment. The new release introduces crucial capabilities for commercial freight operations including night driving and construction zone handling.
By boosting the speed of its AI model training by 10X, SuperDrive 6.0 also enables rapid feature deployment and validation. Built on over 7 million miles of real-world driving across the US, Europe, and Asia, SuperDrive 6.0 is a major advancement that moves the company closer to scalable fully driverless commercial deployment targeted for 2027 with factory built trucks.
This announcement comes as PlusAI plans to become publicly listed via a business combination with Churchill Capital Corp IX.
David Liu, CEO and co-founder of PlusAI, says: “SuperDrive 6.0 isn’t an incremental update; it’s a major advancement of what an autonomous ‘brain’ can do. By adding night driving and construction zone handling, autonomous trucks with SuperDrive could achieve 24/7 commercial operations.
“By doubling our predictive accuracy and supercharging our AI development pipeline, we have the potential to accelerate our development and expand our operating environment faster. From internal validation of a major feature to its deployment into commercial freight operations can take just weeks.”
Operating in strategic freight corridors of Texas
Autonomous trucks integrated with SuperDrive 6.0 are already moving commercial freight in Texas. In addition to overall performance boost, the construction zone handling capability on SuperDrive 6.0 is active, and we expect to release night driving in the coming weeks for 24/7 commercial freight operation on customer routes.
Being able to handle construction zones and night driving helps to enhance road safety and efficiency. Thousands of traffic accidents occur in Texas construction zones each year due to speeding, driver inattention and unsafe lane changes. Additionally, overnight 24/7 commercial freight deliveries could potentially more than double the utilization of a truck.
A step-change in AI economics by accelerating speed and scalability
SuperDrive 6.0 delivers significant improvements in the cost structure of autonomous vehicle development, an area of critical importance to long-term unit economics. Through a combination of autolabeling, imitation learning, and reinforcement learning, PlusAI has achieved a 10X improvement in AI training speed and a 3X reduction in data labeling costs.
These gains enable faster deployment of new features, operating environments, and new routes that SuperDrive can operate in. The tighter cycle from simulation to on-road validation and commercial operation directly compresses time-to-revenue and improves scalability.
Production-grade architecture built for reliability
SuperDrive 6.0 introduces a distributed compute architecture across multi-node, high-performance system-on-chips, including Nvidia DRIVE Orin and Thor, engineered to meet the uptime and efficiency needs of commercial freight operations.
The system maintains performance under sensor degradation, calibration drift, and partial hardware failure: conditions that define the difference between a demonstration system and a commercially deployable product.
At the core of the SuperDrive system is a new Transformer-based “Reflex” layer in the PlusAI AV 2.0 architecture which combines large-scale perception and advanced motion forecasting.
The motion forecasting sub-model achieves a 2x improvement in predicting the trajectories of dynamic actors such as merging vehicles, pedestrians, and lane-changing traffic, leading to safer and smoother real-time decision-making.
Clear path to targeted 2027 scalable driverless commercial launch
SuperDrive 6.0 moves PlusAI a step closer to the planned fully driverless commercial operations launch using factory-built autonomous trucks in 2027. In addition to unlocking important new features of night driving and construction zone handling, the system enables PlusAI to rapidly release new features and new operating environments and routes in the future.
