Lane-Level Route Planning With Probabilistic Lane Transitions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional route planning systems lack lane-level information, leading to unsafe maneuvers and inefficiencies in autonomous driving, as they do not account for lane-specific decisions and contingencies, focusing solely on time minimization without considering user preferences or vehicle competence.
Innovation Solution
A method and system that convert lane segment-level traversal information into probabilities for a state transition function, integrating user preferences and multiple objectives, including slack time, to autonomously control vehicle movements between neighboring lane segments based on current positions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If road-level route planning is used, then the planning complexity is reduced, but the safety and suitability for autonomous driving deteriorates
Solution Approach 1:
The patent segments the route planning into two distinct levels: road-level planning (for overall route determination) and lane-level planning (for specific lane selection and maneuver execution). This segmentation allows the system to maintain simplified high-level planning while implementing detailed lane-level control, resolving the contradiction between planning complexity and autonomous driving safety.
Solution Approach 2:
The patent introduces a new dimension of planning granularity by adding lane-level planning beneath the existing road-level planning. This dimensional extension enables the system to operate effectively at both abstraction levels, maintaining computational efficiency at the road level while achieving the safety and precision required for autonomous driving at the lane level.
2Productivity
If deterministic road-level planning is used, then the route time minimization is achieved, but the adaptability to lane-specific conditions deteriorates
Solution Approach 1:
The patent implements dynamic adaptability at the lane level by allowing the planning system to adjust to real-time lane-specific conditions such as traffic patterns, lane closures, and maneuverability constraints. This dynamic adjustment occurs within the lane-level planning layer, which can modify lane selections and timing based on current conditions while maintaining the overall route time optimization established at the road level.
3Ease of operation
If traditional route planners with distance warnings are used, then the notification simplicity is maintained, but the maneuver safety deteriorates
Solution Approach 1:
The patent introduces an intermediary lane-level planning system that sits between the simple distance warnings and the autonomous vehicle control. This intermediary layer processes the distance warnings and translates them into specific, safe lane-level maneuvers by considering current lane position, available lanes, and optimal transition paths, thereby maintaining notification simplicity while ensuring maneuver safety.
Data Source
AI summary
Lane segment-level traversal information is obtained. The lane segment-level traversal information is converted into probabilities for a state transition function. A policy is derived from a decision model using the state transition function. The policy directs vehicle movement of a vehicle between neighboring lane segments based on a cost function integrating a user preference with respect to at least two objectives and a slack time for alternative routes. The slack time indicates an allowable deviation in travel time relative to the user preference. A destination is received. The vehicle is then autonomously controlled on a route to the destination using the policy for lane transitions based on current lane positions.