Autonomous Driving Trajectory Integration for Collision Mitigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Collision mitigation systems in vehicles face challenges in accurately predicting driver intentions, leading to false positives and increased latency in autonomous driving scenarios, especially at higher speeds, due to ambiguity in situational conditions and the need to interpret human driver inputs.
Innovation Solution
Integrating a lower latency driver assistance system with autonomous driving systems to operate in concert, using precise trajectory information from autonomous driving to reduce ambiguity and enhance collision mitigation, and sharing mitigation measures between systems to improve reaction time and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If collision mitigation systems interpret driver inputs to predict vehicle trajectory, then the system can respond to driver intentions, but false positives increase and reaction latency increases due to situational ambiguity
Solution Approach 1:
The patent introduces an intermediary system (autonomous driving system) that acts as a mediator between the driver and the collision mitigation system. This intermediary provides explicit trajectory information to the collision mitigation system, eliminating the need for the collision mitigation system to interpret driver inputs directly. This resolves the contradiction by maintaining adaptability to driver intentions through the intermediary while improving reliability by removing situational ambiguity in trajectory prediction
Solution Approach 2:
The patent segments the vehicle control system into distinct functional modules: the autonomous driving system responsible for trajectory determination and the collision mitigation system responsible for collision response. By separating these functions, each system can operate with its own optimized logic - the autonomous driving system handles high-level decision making while the collision mitigation system focuses on rapid response using explicit trajectory data, thereby reducing false positives while maintaining responsiveness
2Reliability
If collision mitigation systems are tuned to be less sensitive to avoid false positives, then false positives decrease, but reaction time increases and false negatives increase
Solution Approach 1:
The autonomous driving system performs preliminary action by determining and communicating the vehicle trajectory to the collision mitigation system in advance. This allows the collision mitigation system to be pre-informed of the intended path, enabling it to tune its sensitivity parameters optimally without fearing false positives from misinterpreting driver inputs. The preliminary provision of trajectory information resolves the trade-off by allowing aggressive tuning for fast response while maintaining high reliability
3Extent of automation
If driver assistance systems operate independently with full sensor processing, then system autonomy is maintained, but processing latency increases
Solution Approach 1:
The patent merges the collision mitigation system with the autonomous driving system by having them share the same sensor inputs and trajectory processing pipeline. The collision mitigation system leverages the autonomous driving system's trajectory determination capabilities rather than independently processing all sensor data. This merging maintains sufficient automation independence for safety-critical functions while dramatically reducing processing latency by eliminating redundant sensor processing
Data Source
AI summary
Systems and methods for controlling a vehicle. The system includes one or more sensors positioned on the vehicle and configured to sense an environment surrounding the vehicle, a collision mitigation subsystem configured to control a braking system of the vehicle, and an autonomous driving subsystem communicatively coupled to the one or more sensors and the collision mitigation subsystem. The autonomous driving subsystem is configured to receive sensor information from the one or more sensors and generate, based on the sensor information, a model of the environment surrounding the vehicle. The autonomous driving subsystem is configured to determine, based on the model of the environment surrounding the vehicle, a plurality of possible trajectories for the vehicle and to select, from the plurality of possible trajectories, a travel path for the vehicle. The autonomous driving subsystem is configured to transmit the travel path to the collision mitigation subsystem.