Autonomous Vehicle Collision Mitigation with Trajectory-Based Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately detecting potential collisions due to reliance on a single set of sensors and motion planning systems, leading to potential false positives and reduced safety and reliability.
Innovation Solution
A collision mitigation system that utilizes motion plan data to determine a region of interest in the surrounding environment, independent of the autonomy system's sensors, to identify and avoid potential collisions by controlling the vehicle's actuators and providing data to other onboard systems for improved accuracy and redundancy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single set of sensors and motion planning system is used for collision detection, then the system complexity is reduced, but the measurement precision and reliability of collision detection deteriorate
Solution Approach 1:
The system divides collision detection into two independent segments: the autonomy system handles motion planning and basic obstacle avoidance, while the collision mitigation system independently performs collision risk assessment and mitigation. This segmentation allows each subsystem to specialize in its function, improving overall detection accuracy without requiring a complete redesign of the entire system.
Solution Approach 2:
The collision mitigation system acts as an intermediary layer between the autonomy system and the vehicle control system. It receives motion plan data from the autonomy system, independently assesses collision risks using separate sensors and processing, and provides mitigation commands when necessary. This intermediary structure adds reliability without fully integrating all sensor resources into a single complex system.
2Area of stationary object
If sensor resources are distributed across the entire surrounding environment, then the coverage area is maximized, but the measurement precision for critical collision areas deteriorates
Solution Approach 1:
The collision mitigation system applies local quality by concentrating sensor resources and computational attention on regions of high collision risk (front and side areas where the autonomous vehicle is moving) rather than uniformly distributing them across the entire surrounding environment. This allows for higher detection precision in critical areas while maintaining adequate coverage elsewhere.
Solution Approach 2:
The system performs preliminary identification of regions of interest based on the autonomous vehicle's motion plan before conducting detailed collision risk assessment. By pre-identifying areas where collisions are most likely to occur based on the planned trajectory and speed, the system can allocate sensor resources more effectively to these critical zones in advance.
3Device complexity
If the autonomy system independently handles all motion planning and obstacle avoidance, then the device complexity is minimized, but the reliability of collision avoidance deteriorates
Solution Approach 1:
The system architecture is segmented into two independent but coordinated subsystems: the autonomy system for motion planning and the collision mitigation system for collision risk assessment. This segmentation creates functional redundancy where each system can independently evaluate collision risks, thereby improving reliability without requiring complete integration of all functions into a single complex system.
Solution Approach 2:
The collision mitigation system provides beforehand cushioning by independently assessing collision risks and preparing mitigation commands in advance. It receives motion plan data from the autonomy system, evaluates potential collision scenarios before they occur, and has pre-prepared mitigation strategies ready to execute, thereby cushioning against potential failures in the autonomy system's collision avoidance.
Data Source
AI summary
Systems and methods for controlling an autonomous vehicle are provided. In one example embodiment, a computer-implemented method includes obtaining, from an autonomy system, data indicative of a planned trajectory of the autonomous vehicle through a surrounding environment. The method includes determining a region of interest in the surrounding environment based at least in part on the planned trajectory. The method includes controlling one or more first sensors to obtain data indicative of the region of interest. The method includes identifying one or more objects in the region of interest, based at least in part on the data obtained by the one or more first sensors. The method includes controlling the autonomous vehicle based at least in part on the one or more objects identified in the region of interest.