Autonomous Junction Maneuvering With Dynamic Roadside Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Merging of vehicles at junctions, particularly at unsignalized junctions with obstructed views, is challenging for both human drivers and autonomous vehicles due to limited sensor fields of view and difficulty in assessing traffic conditions.
Innovation Solution
The method involves a partially autonomous vehicle observing a first area of a second road, altering its detection area to include a second, previously unseen area, and then maneuvering into the second road based on observations from the expanded detection area.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the vehicle uses a fixed detection area for observing the second road, then the detection system can maintain stable operation, but the field of view remains limited and cannot detect road users in obscured areas
Solution Approach 1:
The detection area is made dynamic by altering its boundaries based on the vehicle's position and orientation at the junction. The system continuously adjusts the detection area to sweep across different portions of the second road, transforming a static detection zone into a dynamic one that adapts to the vehicle's maneuvering state, thereby expanding the overall field of view while maintaining reliable detection through systematic coverage
Solution Approach 2:
The patent introduces temporal dimension to the detection area by observing different areas at different times during the maneuvering process. Instead of a single static detection zone, the system creates a sequence of detection areas that collectively cover a larger spatial volume, effectively adding a time dimension to expand the observable area without compromising detection stability at any moment
2Area of stationary object
If the vehicle alters the detection area to include previously unseen areas, then the field of view increases and more road users can be detected, but the detection system complexity increases
Solution Approach 1:
The detection area is segmented into multiple discrete portions that are observed sequentially during the maneuvering process. Instead of attempting to observe the entire second road simultaneously with a complex wide-angle system, the patent divides the detection task into segments corresponding to different spatial zones, each observed by the existing detection system at different times, reducing overall system complexity while achieving comprehensive coverage
Solution Approach 2:
The detection system performs preliminary observation of the first area before altering to observe the second area. This sequential preliminary action allows the system to build up information about traffic conditions progressively, simplifying the detection process by breaking it into preparatory stages rather than requiring simultaneous complex analysis of all areas
Data Source
AI summary
The disclosure relates to a method for maneuvering an at least partially autonomous vehicle from a first road positioned at a junction into a second road meeting the first road at the junction, the method comprising: observing a first area of the second road for detection of road users by a detection system of the vehicle; altering a detection area of the detection system to include a second area of the second road, the second area comprising a further area of the second road that is not comprised in the first area; observing the second area of the second road for detection of the road users by the detection system; and maneuvering the vehicle into the second road based on the observing of the second area of the second road.