Autonomous Overtaking Control With Lead Vehicle Intention Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for autonomous overtaking on two-lane roads do not effectively account for the reaction of other vehicles to the ego vehicle's overtaking maneuver, particularly in the presence of an oncoming vehicle.
Innovation Solution
A method that estimates the intention of the lead vehicle using data from sensors and performs a vehicle maneuver based on this intention, while also considering the presence of an oncoming vehicle to ensure safe overtaking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If autonomous overtaking is performed without considering the reaction of other vehicles, then the overtaking maneuver can be executed more quickly and efficiently, but the safety of the maneuver is compromised due to inability to predict lead vehicle reactions
Solution Approach 1:
The system performs preliminary actions by estimating the lead vehicle's intention before executing the overtaking maneuver. Multiple potential intention models are developed and evaluated in advance, allowing the autonomous vehicle to predict lead vehicle reactions and plan accordingly, rather than reacting after the fact
Solution Approach 2:
An intention estimation system acts as an intermediary between the autonomous vehicle's overtaking decision and the actual maneuver execution. This intermediary layer processes sensor data, evaluates multiple intention models, and provides predicted lead vehicle responses that inform the overtaking control, bridging the gap between simple overtaking commands and complex real-world interactions
2Reliability
If the autonomous vehicle waits for complete information about lead vehicle intention before overtaking, then safety is improved, but the overtaking time increases significantly
Solution Approach 1:
The system applies partial action by using a set of predefined intention models that cover the most likely lead vehicle behaviors without requiring complete certainty about the actual intention. Rather than waiting for full information, the system uses these partial models to make sufficiently safe decisions in a timely manner
Solution Approach 2:
The system changes parameters by evaluating multiple intention models with different assumptions about lead vehicle behavior (e.g., cooperative vs. non-cooperative, maintaining speed vs. accelerating). By switching between different model parameters based on observed behavior, the system can make rapid safety assessments without waiting for complete information
3Measurement precision
If multiple intention models are evaluated to predict lead vehicle reaction, then the accuracy of safety assessment is improved, but the computational load increases
Solution Approach 1:
The intention estimation problem is segmented into multiple discrete, pre-defined intention models rather than attempting a single comprehensive analysis. Each model represents a specific lead vehicle behavior pattern, allowing the system to evaluate multiple scenarios independently and efficiently select the most relevant ones for prediction
Data Source
AI summary
The present disclosure provides a method in a data processing system that includes at least one processor and at least one memory. The at least one memory includes instructions executed by the at least one processor to implement a vehicle overtaking system. The method includes causing a vehicle control system in an ego vehicle to execute at least a portion of an input control sequence, receiving, from a first plurality of sensors coupled to the ego vehicle, lead vehicle data about a lead vehicle, estimating an intention of the lead vehicle based on the lead vehicle data, and causing the vehicle control system to perform a vehicle maneuver based on the intention of the lead vehicle.