Autonomous Lane Change Feasibility Using Limit Trajectories
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current lane change assistance systems for autonomous vehicles are resource-intensive and struggle to anticipate events effectively, making lane changes risky due to their inability to consider multiple trajectories and dynamic environmental changes.
Innovation Solution
A method to determine the feasibility percentage of a lane change by acquiring target inter-vehicular time, destination time, and longitudinal dynamics of the ego-vehicle and adjacent objects, calculating limit trajectories, and assessing the percentage of feasible trajectories, which is dynamic and resource-efficient, allowing for periodic updates and consideration of future object locations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current lane change assistance systems use powerful processors and large memory to process data in real time, then the system can handle complex lane change scenarios, but the device complexity and cost increase significantly
Solution Approach 1:
The patent segments the lane change assessment into distinct phases: identifying candidate trajectories, evaluating constraints for each trajectory, calculating feasibility percentages, and selecting the optimal trajectory. This segmentation allows the system to process complex lane change scenarios through a series of simpler, more manageable computational steps, reducing the burden on processors and memory while maintaining high reliability
Solution Approach 2:
The system performs preliminary actions by pre-identifying candidate trajectories and pre-evaluating constraints before the actual lane change execution. By calculating feasibility percentages in advance for multiple potential trajectories and filtering them based on constraints, the system prepares the optimal path beforehand, reducing real-time computational complexity while ensuring safe lane changes
2Speed
If current systems initiate lane change as soon as there is only one possible trajectory, then the response time is fast, but the system cannot anticipate future events throughout the entire maneuver duration
Solution Approach 1:
The system performs preliminary evaluation of multiple trajectories and their feasibility percentages before initiating the lane change. By assessing candidate trajectories in advance and selecting the optimal one based on feasibility analysis, the system ensures both rapid response and the ability to anticipate future events throughout the maneuver
Solution Approach 2:
The system dynamically adjusts the lane change initiation decision based on real-time feasibility assessments. By continuously evaluating the feasibility percentage of candidate trajectories and adapting to changing environmental conditions, the system can initiate lane changes rapidly when safe while maintaining the capability to anticipate and respond to future events throughout the maneuver duration
3Adaptability or versatility
If the system calculates feasibility percentage periodically and updates destination time dynamically, then the system adapts to environmental changes, but the computational load increases
Solution Approach 1:
The system implements periodic calculation of feasibility percentages at predetermined time intervals during the lane change maneuver. This periodic approach allows the system to adapt to environmental changes systematically while managing computational energy consumption by performing assessments at regular intervals rather than continuously, balancing adaptability with energy efficiency
Solution Approach 2:
The system dynamically updates the destination time based on the current phase of the lane change maneuver and environmental conditions. By adjusting the timing and focus of feasibility calculations dynamically rather than using a fixed approach, the system adapts to changing conditions while optimizing computational energy consumption by concentrating calculations when most critical
Data Source
Figure 1
Figure 2
AI summary
The invention relates to a method and a device for determining a feasibility percentage of a lane change for an autonomous vehicle, referred to as an ego-vehicle, travelling on an ego-lane. The method comprises the steps of acquiring (201) a target inter-vehicular time, acquiring (202) a target lane, acquiring (203) a destination time, and acquiring (204 and 205) characteristics of the longitudinal dynamics of the ego-vehicle and adjacent objects. For each object, a limit path of the ego-vehicle is determined (206). A feasibility percentage is determined (207) from the set of limit paths and a set of possible paths for the ego-vehicle, the set of possible paths for the ego-vehicle being determined from the characteristics of the longitudinal dynamics of the ego-vehicle.