Off-road machine-learned obstacle navigation in an autonomous vehicle environment

This invention describes an autonomous off-road vehicle designed to navigate a route. When it encounters an obstruction, it first uses a machine-learned model, fed by camera images and depth data, to identify what it is. If that model cannot identify the obstruction, a second machine-learned model decides if the obstruction is small enough to simply ignore. If the obstruction cannot be ignored, the vehicle will then take action, such as changing its path or alerting a human operator.

Why it matters: Since 2020, machine learning models for real-time object detection and classification have become significantly more robust and efficient, making the dual-model approach more feasible. Advances in sensor fusion also simplify integrating camera and depth data for complex off-road environments.

Status: Not in forceListed as no longer active. The specific reason is not in the record we hold.
How hard to build: SpecializedAutonomous vehicle, machine learning models, sensor fusion

3 AI reinvention directions for this patent

See all three

AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.

Reinvent this with AI

Where this one came from

Every morning we go through millions of abandoned patents and pick out 15 worth your time. This was one of today’s.

This one

+10