How Autonomous Trucks Predict and Handle Tire Blowouts

This patent describes how an autonomous vehicle uses real-time sensor data, road conditions, and a dynamic model to predict tire failure and take corrective actions like adjusting its path or pulling over.

What it covers

This patent outlines a method for an autonomous vehicle to evaluate its tires and prevent or handle failures. First, the vehicle's processors obtain baseline information about its tires (Claim 1). While driving, it receives real-time sensor data, such as tire pressure, temperature, or shape (Claim 5), potentially from a tire pressure monitoring system (Claim 6) or even cameras and lidar (Claim 7). This data updates a 'dynamics model' that represents how the tires behave (Claim 1). The system also considers external factors like roadway conditions (e.g., obstacles like potholes (Claim 2, 3)) or environmental conditions (e.g., ambient temperature (Claim 4)). Based on the updated model and these conditions, the processors determine if the chance of a tire failure, such as a blowout, slow leak, or tread damage (Claim 10), exceeds a set limit. If it does, the autonomous vehicle takes a corrective action. For example, if a pothole is detected ahead and the system predicts a high risk of tire damage, the vehicle might adjust its position within the lane to avoid the impact (Claim 9). Other actions could include pulling over, changing its route, or notifying a remote service (Claim 11).

What it doesn't cover

• —Does not cover tire monitoring systems in vehicles that are solely human-driven, as the claims specifically refer to an 'autonomous vehicle'.
• —Does not cover systems that only provide a simple alert (e.g., 'low tire pressure') without updating a 'dynamics model' to predict future failure based on multiple factors.
• —Does not cover systems where a human driver makes all decisions for corrective action, as the claims state the processors 'causing the autonomous vehicle to take a corrective action'.
• —Does not cover systems that only use internal tire sensor data without also considering external 'roadway condition' or 'environmental condition' information.
• —Does not cover a system that only detects a tire failure after it has occurred without attempting to predict the possibility beforehand.

The clever bit

The cleverness lies in combining diverse data sources—baseline tire characteristics, real-time tire sensor readings, and predictive environmental/roadway conditions—into a dynamic model. This allows the autonomous vehicle to proactively assess the risk of tire failure and take specific, automated corrective actions *before* a problem escalates or even occurs, rather than just reacting to an existing failure.

Why it matters

Tire blowouts on large trucks can be extremely dangerous, leading to accidents and significant downtime. For autonomous trucks, predicting and handling these events without human intervention is critical for safety, reliability, and public trust. This technology helps ensure that self-driving vehicles can operate safely and efficiently, even when facing potential mechanical issues or challenging road conditions.

Real-world examples

• 1.Waymo autonomous trucks
• 2.Aurora Innovation self-driving trucks
• 3.TuSimple autonomous freight solutions
• 4.Future autonomous commercial delivery fleets