Autonomous Vehicle Trajectory Control With Local Map Adaptation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle control technologies are limited to predefined traffic scenarios and struggle with adapting to dynamic changes in the environment, leading to inefficiencies and reduced responsiveness in real-time driving situations.
Innovation Solution
A method and device for controlling autonomous vehicles that utilize perception means to detect real-time situations and optimize trajectories by modifying local maps and kinematic profiles, allowing for increased responsiveness and adaptability to dynamic traffic conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If predefined trajectories from high-definition maps are used, then the autonomous vehicle can navigate structured traffic zones, but the system is intrinsically limited to predefined traffic scenarios and cannot adapt to dynamic changes
Solution Approach 1:
The system dynamically adapts the predefined trajectory by detecting deviations caused by dynamic obstacles (pedestrians, animals, vehicles) and computing corrected trajectories in real-time. The perception system continuously monitors the environment and updates the trajectory parameters to maintain safety while following the planned path, transforming a static map-based approach into a dynamic adaptive system.
Solution Approach 2:
The system implements feedback by comparing the detected environment with the predefined map data, identifying situations to be optimized, and adjusting the trajectory accordingly. The perception system provides continuous feedback about dynamic obstacles, and the control system uses this feedback to modify speed profiles and trajectory parameters, creating a closed-loop adaptive navigation system.
2Stability of the object's composition
If the autonomous vehicle strictly follows predefined speed profiles, then it maintains predictable motion patterns, but it cannot respond quickly to unexpected situations in the environment
Solution Approach 1:
The speed profile is transformed from a static predefined parameter into a dynamic variable that adjusts in real-time based on detected situations. When dynamic obstacles are detected, the system computes optimized speed profiles that maintain stability for normal operation but enable rapid response when needed, allowing the vehicle to accelerate or decelerate dynamically while maintaining overall motion control.
Solution Approach 2:
The system changes speed profile parameters dynamically by computing maximum collision-free speed thresholds based on detected obstacle distances and relative velocities. The speed parameters are adjusted as functions of perception data, allowing the vehicle to maintain stable cruising speeds under normal conditions while rapidly adapting speed when dynamic situations arise.
3Device complexity
If the autonomous vehicle uses a single global trajectory, then it maintains simple control logic, but it cannot optimize for local traffic conditions and situations
Solution Approach 1:
The trajectory is segmented into multiple sections, with each segment optimized for specific local conditions. The system divides the global route into manageable portions and applies local optimizations to each segment based on detected situations, maintaining simple control logic for each segment while achieving comprehensive adaptability across the entire journey.
Solution Approach 2:
The system applies local quality by optimizing trajectory and speed parameters specifically for detected local situations rather than using uniform global parameters. When situations to be optimized are detected (such as pedestrians, animals, or unusual traffic patterns), the system computes localized trajectory corrections and speed adjustments that are applied only to affected segments, maintaining simplicity elsewhere.
Data Source
AI summary
A method controls an autonomous vehicle equipped with at least one perception unit and a first map including stored digital data representing the actual infrastructure of the environment of the autonomous vehicle. The method includes defining a first trajectory of the autonomous vehicle and a first associated kinematic profile, detecting, from data transmitted by the perception unit, a situation to be optimized on a given segment of the first trajectory located in front of the autonomous vehicle, optimizing at least one element from the first map and the first kinematic profile, optimizing the first map including determining a local map on the given segment, and/or optimizing the first kinematic profile including a determination of a second kinematic profile associated with the given segment, and controlling the displacement of the autonomous vehicle over the given segment taking into account the local map and/or the second kinematic profile.