Atieva, Inc. has secured a major milestone in the Automotive, Battery and Self-Driving Technologies industry with a newly patented system for smooth low-speed stops. This innovation focuses on a newly awarded patent, titled ‘Smooth automated braking at low speed’, which won Swanson Reed’s Patent of the Month for January 2026 for being an outstanding invention in its field. The patent describes a sophisticated method for dynamically calculating and applying braking profiles to ensure smooth stops without abrupt jolts.

Overcoming Abrupt Low-Speed Braking

Specifically, the abstract outlines that braking of a vehicle at low speed comprises: dynamically determining, while a vehicle is controlled by a driver, a deceleration limit for the vehicle with regard to an obstacle, the deceleration limit determined based on at least (i) the low speed, (ii) a distance to the obstacle, and (iii) a determined jerk limit for the vehicle; after dynamically determining the deceleration limit, determining regions to include in a braking profile for the vehicle; when the regions determined do not include a constant deceleration region, again determining the deceleration limit, wherein the deceleration limit is not again determined when the regions determined do include the constant deceleration region; determining whether to brake the vehicle, the determination based on at least the low speed, the deceleration limit, and the distance to the obstacle; and in response to a determination to brake the vehicle, braking the vehicle according to the braking profile.

Meeting the U.S. R&D Tax Credit Rules

To qualify for the Research and Development (R&D) tax credit under IRC Section 41, this patented technology successfully aligns with the IRS Four-Part Test:

• Permitted Purpose: The development aims to create a new or improved business component—specifically, enhancing the performance and quality of automated braking systems to reduce “jerk” and improve passenger comfort.
• Technological in Nature: The innovation fundamentally relies on the hard sciences of computer science, physics, and automotive engineering to dynamically calculate deceleration limits based on speed, distance, and jerk constraints.
• Elimination of Uncertainty: At the project’s outset, engineers faced technical uncertainty regarding the optimal algorithmic methodology needed to reliably predict and execute smooth braking profiles in unpredictable, real-world low-speed scenarios.
• Process of Experimentation: Developing this dynamic profile required extensive systematic trial and error, including software simulations, algorithm modeling, and real-world track testing to validate the braking profile’s effectiveness across different regions of deceleration.

Three Practical Applications Qualifying for R&D Tax Credits

1. Advanced Stop-and-Go ADAS (Advanced Driver Assistance Systems): Integrating this logic into an ADAS designed for heavy, bumper-to-bumper traffic. Qualifying R&D activities would involve experimenting with sensor fusion (radar and cameras) and iterative algorithm testing to eliminate the uncertainty of safely applying the braking profile without causing motion sickness for occupants.
2. Automated Self-Parking Systems: Adapting the dynamic deceleration methodology for parallel and perpendicular autonomous parking. The R&D process would require evaluating spatial constraints, systematically testing different vehicle weights, and refining jerk limits to prevent jarring halts against curbs or walls—satisfying the requirement for a process of experimentation.
3. Autonomous Last-Mile Delivery Vehicles: Calibrating the low-speed braking profiles for unmanned delivery bots operating in dense pedestrian zones. Qualifying work involves systematic trial and error to adjust the deceleration limits based on varying payload weights to protect fragile cargo while ensuring immediate, smooth stopping to avoid sudden pedestrian obstacles.