Autonomous Brake Preload Control for Low-Confidence Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in ensuring safe operation due to perception inaccuracies, leading to unnecessary braking and increased stopping distances, as they struggle to accurately detect and predict objects, especially at lower confidence levels or with spurious sensor noise.
Innovation Solution
The implementation of a pre-loading mechanism for the braking system, where the system is partially pressurized in anticipation of potential braking events based on perception accuracy levels, allowing additional time for improved object recognition and reducing the need for excessive braking distances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the autonomous vehicle applies brakes immediately upon detecting an object with low confidence, then safety is improved, but unnecessary braking occurs increasing stopping distances and reducing ride comfort
Solution Approach 1:
The braking system is pre-loaded in advance when object detection confidence is below a threshold, preparing the brakes for immediate application without actually applying braking force yet. This preliminary action reduces the effective stopping distance while avoiding unnecessary braking by maintaining the pre-loaded state only temporarily.
Solution Approach 2:
The braking system transitions between three dynamic states: quiescent (normal), pre-loaded (intermediate), and applied (braking). This dynamic state management allows the system to adapt its readiness level based on perception confidence, enabling faster response when needed while avoiding unnecessary braking in uncertain situations.
2Loss of energy
If the autonomous vehicle waits for higher confidence in object detection, then unnecessary braking is reduced, but response time increases compromising safety
Solution Approach 1:
The system performs preliminary preparation by pre-loading the brakes when confidence is low, rather than waiting for high confidence or applying brakes immediately. This intermediate preliminary action reduces energy consumption compared to continuous braking while maintaining safety readiness.
Solution Approach 2:
The system continuously monitors perception confidence levels and adjusts braking system state accordingly. When confidence improves above the threshold during the pre-loaded state, the system cancels the pre-load and avoids braking, using feedback to optimize the balance between safety and energy consumption.
3Speed
If the braking system is constantly pre-loaded, then braking response time is improved, but wear and energy consumption increase
Solution Approach 1:
Instead of constant pre-loading, the system applies preliminary action selectively only when perception confidence is below the threshold. This conditional preliminary action maintains fast response capability when needed while avoiding unnecessary energy consumption during normal high-confidence operation.
Solution Approach 2:
The braking system alternates between quiescent and pre-loaded states based on periodic perception confidence assessments. This periodic switching ensures fast response is available only during periods of uncertainty, optimizing the balance between response speed and energy consumption.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach reduces braking distances by up to 30 feet when stopping from 60 miles per hour, enhances ride comfort, and maintains safety by enabling more immediate braking force application when necessary, while minimizing wear and energy consumption.
Implementation Method 1
The braking system includes a compressor that provides pressurized air to a reservoir
Implementation Method 2
a service brake valve on the service brake line to control a flow of pressurized air from the pressurized air reservoir to the service brake chamber
Implementation Method 3
pre-loading a braking system of an autonomous vehicle... where the system is partially pressurized in anticipation of potential braking events
Data Source
AI summary
Vehicles according to at least some embodiments of the disclosure include a sensor, and a computing device comprising at least one hardware processing unit. The computing device is programmed to perform operations comprising capturing an image with the sensor, identifying an object in the image, and in response to an accuracy of the identification meeting a first criterion, pre-loading a braking system of the autonomous vehicle. In some aspects, the computing device may predict that an object not currently within a path of the vehicle has a probability of entering the path of the vehicle that meets a second criterion. When the probability of entering the path meets the second criterion, some of the disclosed embodiments may pre-load the braking system.