Autonomous Lane Change Decision via Obstacle Trajectory Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional driverless vehicles lack the ability to respond effectively to emergencies during the lane changing process due to inadequate assessment of obstacle movements and lane safety.
Innovation Solution
A decision method and device that acquire planned tracks for a driverless vehicle to switch lanes, predict obstacle movements, and determine safe travel motions by calculating distances and safety thresholds, allowing the vehicle to adjust its lane changing process accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the driverless vehicle uses conventional lane changing control without obstacle prediction, then the control system is simple, but the vehicle cannot respond to emergencies during lane changing
Solution Approach 1:
The system performs preliminary prediction of obstacle trajectories before the lane changing maneuver is completed. By predicting where obstacles will be in the future based on current motion states, the system prepares safety assessment information in advance, enabling emergency response without requiring complex real-time reaction mechanisms.
Solution Approach 2:
The patent implements a safety buffer mechanism by calculating predicted tracks of obstacles and comparing them with the planned lane changing trajectory. This creates a protective assessment layer that identifies potential collisions before they occur, allowing the vehicle to adjust or abort the lane changing maneuver proactively rather than reactively.
2Reliability
If the driverless vehicle predicts obstacle tracks and calculates safety distances, then collision avoidance capability is improved, but computational complexity increases
Solution Approach 1:
The prediction system uses the obstacle's own current motion state (velocity, acceleration, heading) to generate its future trajectory. Rather than requiring complex external sensing or prediction models, the system leverages the obstacle's self-motion characteristics to forecast its path, reducing computational burden while maintaining accuracy.
Solution Approach 2:
The patent transforms the complex problem of collision avoidance into a parameter comparison task. By calculating key parameters such as predicted position, distance to vehicle, and time-to-collision, the system reduces the dimensionality of the safety assessment problem, making it computationally tractable while preserving essential safety information.
3Reliability
If the driverless vehicle continuously monitors and re-evaluates lane safety during the lane changing process, then safety is improved, but response time is reduced
Solution Approach 1:
The system implements periodic re-evaluation of lane safety at predetermined intervals or at key milestones during the lane changing process. Rather than continuous monitoring, the vehicle assesses safety at structured checkpoints (e.g., before lane change initiation, during mid-transition, near completion), balancing safety assurance with efficient use of decision time.
Solution Approach 2:
The patent performs preliminary safety assessment by predicting obstacle trajectories in advance and evaluating potential conflicts before they materialize. This upfront evaluation reduces the need for time-consuming real-time analysis during critical moments of the lane changing maneuver, as much of the safety verification is completed beforehand.
Data Source
AI summary
A decision method, device, equipment in a lane changing process and storage medium are provided. The method includes: acquiring a first planned track of a driverless vehicle for travelling to a first lane and a second planned track of the driverless vehicle for travelling to a second lane within a preset time period, in a lane changing process of the driverless vehicle; predicting a predicted track of at least one obstacle within the preset time period according to a travelling state of the obstacle, wherein the obstacle is in a preset range around the driverless vehicle; and determining a travelling motion of the driverless vehicle according to the first planned track, the second planned track and the predicted track of the obstacle.