Autonomous Emergency Braking Using Object Confidence Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles often perform unnecessary and potentially dangerous hard braking due to low-confidence object detections, increasing collision risk and decreasing passenger comfort.
Innovation Solution
The vehicle's computing devices prioritize emergency braking only for high-confidence object detections, using confidence values and object types to determine priority levels, and apply maximum braking levels only when necessary to ensure safety and comfort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the vehicle applies maximum braking level for all detected objects, then safety is improved, but passenger comfort deteriorates due to unnecessary hard braking
Solution Approach 1:
The patent applies different braking strategies based on the specific characteristics of each detected object. High-confidence objects trigger maximum braking, while low-confidence objects use gentle or no braking. This local differentiation of response quality resolves the contradiction by applying strong braking only where necessary for safety while maintaining comfort for non-critical detections.
Solution Approach 2:
The system changes the braking parameter (braking level) based on the confidence value of object detection. When confidence is high, maximum braking is applied; when confidence is low, reduced or no braking is applied. This dynamic parameter adjustment resolves the contradiction between safety and comfort by adapting the braking intensity to the reliability of the detection.
2Speed
If the vehicle uses low-confidence object detections for emergency braking, then response speed is improved, but reliability deteriorates due to false positives
Solution Approach 1:
The system performs preliminary assessment of detection confidence before initiating emergency braking. By evaluating the confidence value in advance, the system filters out low-confidence false positives while maintaining rapid response to high-confidence threats. This preliminary filtering action resolves the contradiction between response speed and reliability.
Solution Approach 2:
The confidence value acts as an intermediary parameter between object detection and braking execution. This intermediary layer filters detections based on reliability, allowing the system to respond quickly to valid threats while ignoring false positives. The confidence metric mediates between detection data and control action, resolving the speed-reliability contradiction.
3Object-affected harmful factors
If the vehicle applies gentle braking for all objects, then passenger comfort is improved, but safety deteriorates due to insufficient braking for real threats
Solution Approach 1:
The braking system dynamically adjusts its response based on the confidence level of object detection. Rather than using a fixed gentle braking level, the system transitions between gentle and maximum braking modes depending on detection reliability. This dynamic adaptation resolves the contradiction by providing comfort for uncertain detections while ensuring safety for confident detections.
Solution Approach 2:
Different braking intensities are applied to different detected objects based on their confidence values. High-confidence objects receive maximum braking for safety, while low-confidence objects receive gentle braking for comfort. This localized quality differentiation resolves the contradiction between comfort and safety by matching braking intensity to detection reliability.
Data Source
AI summary
Aspects of the disclosure provide for generation of trajectories for a vehicle driving in an autonomous driving mode. For instance, information identifying a plurality of objects in the vehicle's environment and a confidence value for each of the objects is received. A set of constraints may be generated. That one or more processors are unable to solve for a trajectory given the set of constraints and an acceptable braking limit may be determined. A first constraint is identified as a constraint for which could not be solved and a first confidence value. That the vehicle should apply a maximum braking level is determined based on the identified first confidence value, a threshold, and the determination that the one or more processors are unable to solve for a trajectory. Based on the determination that the vehicle should apply the maximum braking level, the maximum braking level is applied.