Platooning Trajectory Control Using Reinforcement Learning Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing platooning systems face challenges in maintaining stable vehicle following and efficient control of rear vehicles based on the traveling trajectory of front vehicles, particularly in exceptional situations.
Innovation Solution
A platooning control device utilizing reinforcement learning with image information and control points to stabilize vehicle following, employing a learning device, compensation determination unit, and inferring neural network to generate steering, braking, and acceleration control signals based on feedback from front vehicle coordinates and radio signal strength.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If reinforcement learning is performed using only basic traveling trajectory data, then the learning process is computationally simple, but the vehicle cannot stably follow the trajectory in exceptional situations
Solution Approach 1:
The patent pre-calculates and stores optimal control signals for various exceptional situations during the reinforcement learning training phase. When exceptional situations occur during actual platooning, the system quickly retrieves pre-computed control strategies rather than performing real-time complex calculations, thereby improving reliability without significantly increasing operational complexity
Solution Approach 2:
The patent introduces an intermediate processing layer that includes compensation determination units and situation judgment units. These intermediaries process raw sensor data and trajectory information before feeding them to the reinforcement learning model, enabling more stable trajectory following in exceptional situations while managing system complexity through modular architecture
2Measurement precision
If detailed image information and control points are used for reinforcement learning, then trajectory following accuracy is improved, but calculation load and data size increase
Solution Approach 1:
The patent extracts only the essential and relevant features from image information, specifically focusing on control points that define the traveling trajectory. By taking out only the necessary trajectory-defining elements rather than processing complete image data, the system achieves high trajectory following accuracy while significantly reducing calculation load and data processing requirements
Solution Approach 2:
The patent applies local quality by focusing computational resources on specific critical regions - the control points along the trajectory - rather than uniformly processing all image data. This localized approach to feature extraction and processing improves trajectory following accuracy where it matters most while minimizing overall calculation load
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A platooning control device includes: a learning device configured to perform reinforcement learning on the basis of image information and a feedback signal and to control a pertinent vehicle so as to follow a traveling trajectory of a front vehicle according to a result of the reinforcement learning; and a compensation determination unit configured to receive a coordinate of a control point regarding the traveling trajectory of the front vehicle from the front vehicle and to compare a coordinate of the pertinent vehicle with the coordinate of the control point, thereby generating the feedback signal.