Autonomous Vehicle Pull-Over Trajectory and Braking Profile
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles lack an effective method to identify safe regions on a road for pulling over and stopping, especially in emergency scenarios, due to limitations in detecting road boundaries and obstacles, which can lead to instability and control issues during braking.
Innovation Solution
A method and system that utilize lane boundaries, road edges, and vehicle size to determine a braking profile and trajectory for an autonomous vehicle to safely pull over and stop, incorporating sensors and map data to identify suitable regions and adjust speed phases to avoid obstacles, ensuring stability and control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If autonomous vehicles use basic braking methods to stop quickly, then stopping time is reduced, but vehicle stability and control are compromised
Solution Approach 1:
The braking process is divided into three distinct phases: pre-maneuver phase (reducing speed at first rate), maneuver phase (reducing speed at second rate), and post-maneuver phase (reducing speed at third rate). This segmentation allows the vehicle to progressively reduce speed while maintaining stability throughout the stopping process, rather than applying maximum braking force immediately which would compromise control.
Solution Approach 2:
The system dynamically adjusts braking force based on the vehicle's current state and phase of operation. The braking profile is not static but adapts in real-time, switching between different braking rates (first rate, second rate, third rate) depending on whether the vehicle is in pre-maneuver, maneuver, or post-maneuver phase, thereby optimizing both stopping time and stability.
2Device complexity
If autonomous vehicles identify pull-over regions using limited sensor data, then system complexity is reduced, but detection accuracy and safety are compromised
Solution Approach 1:
The system merges multiple data sources including lane boundary detection, road boundary identification, obstacle detection, and map data into a unified pull-over region identification process. By combining these different types of information, the system achieves high detection accuracy without requiring each individual sensor or processing module to be overly complex.
Solution Approach 2:
The computing device performs multiple functions using a single integrated system: it detects lane boundaries, identifies road edges, locates obstacles, determines suitable pull-over regions, and calculates braking profiles. This multi-functional approach reduces overall system complexity compared to having separate dedicated systems for each function.
3Productivity
If autonomous vehicles use aggressive braking to reduce speed rapidly, then productivity is improved, but vehicle control and safety are compromised
Solution Approach 1:
The system performs preliminary speed reduction in the pre-maneuver phase before executing the main pulling over maneuver. By reducing speed at a first rate during this initial phase, the vehicle prepares for the subsequent maneuver phase, ensuring that aggressive braking is only applied when the vehicle is already in a controlled state, thus maintaining reliability while improving emergency response efficiency.
Solution Approach 2:
The braking system changes operational parameters (braking force magnitude) based on the phase of the maneuver. The system transitions between different braking rates (first rate < second rate < third rate) depending on whether the vehicle is in pre-maneuver, maneuver, or post-maneuver phase. This parameter adjustment allows aggressive braking to be applied selectively during the maneuver phase when control is already established, rather than continuously which would compromise reliability.
Data Source
AI summary
Methods and systems for determining instructions for pulling over an autonomous vehicle are described. An example method may involve identifying a region of a road ahead of the autonomous vehicle based on lane boundaries of the road, one or more road boundaries indicating an edge of the road, and a size of the autonomous vehicle. The method may also involve determining a braking profile for reducing the speed of the autonomous vehicle based on the region and a speed of the autonomous vehicle. The method may also involve determining, based on the braking profile, a trajectory such that the autonomous vehicle will travel within the region while reducing the speed of the autonomous vehicle. The method may further involve determining instructions for pulling over and stopping the autonomous vehicle in the region in accordance with the determined trajectory and storing the instructions in a memory accessible by a computing device.