Driving Route Planning With Personalized Style Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional motion trajectory planning in autonomous vehicles is inaccurate due to uniform driving rules, leading to suboptimal driving experiences and safety issues in real-world scenarios with uncertainties.
Innovation Solution
A driving route planning method that utilizes a target route planning model and a target driving style model to analyze user-specific driving data, incorporating LSTM, RNN, and CNN models to generate routes aligned with individual driving habits, using autoencoders to filter noise and optimize routes based on user preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional uniform driving rules are used for motion trajectory planning, then the planning process is simple and fast, but the route planning accuracy is poor and cannot adapt to real-world uncertainties
Solution Approach 1:
The patent applies dynamics by transitioning from static uniform driving rules to a dynamic planning system that adapts to real-world uncertainties. The motion trajectory planning network dynamically adjusts routes based on live sensor data, obstacle positions, and environmental conditions, enabling the system to respond flexibly to changing scenarios while maintaining high planning accuracy.
Solution Approach 2:
The patent implements feedback mechanisms where the planning system continuously monitors actual driving conditions and adjusts trajectories accordingly. Sensor feedback about obstacles, road conditions, and vehicle state is integrated into the planning loop, allowing the system to learn from real-time data and improve route accuracy without requiring complex manual programming.
2Adaptability or versatility
If traditional uniform driving rules are used, then the system is easy to implement, but it cannot adapt to individual driving styles and real-world uncertainties
Solution Approach 1:
The patent applies self-service by enabling the planning system to automatically learn and adapt to individual driving styles through training on historical driving data. The motion trajectory planning network self-adjusts to match driver preferences and behavioral patterns without requiring manual reconfiguration, making the system both highly adaptable and relatively easy to implement through automated learning processes.
Solution Approach 2:
The patent utilizes parameter changes by modifying planning parameters based on learned driving styles and environmental conditions. The system adjusts trajectory parameters such as speed, lane changes, and turning angles according to the driver's preferred behavior patterns, enabling adaptation to individual driving styles while maintaining a relatively simple base implementation framework.
3Reliability
If simple uniform rules are applied, then computational requirements are low, but routing precision and safety are compromised in uncertain environments
Solution Approach 1:
The patent applies preliminary action by pre-training the motion trajectory planning network on extensive historical driving data and simulated scenarios before actual deployment. This preliminary learning phase enables the system to handle uncertain situations more reliably during runtime with lower computational overhead, as the complex decision-making logic is pre-computed and stored in the trained network models.
Solution Approach 2:
The patent replaces mechanical rule-based decision systems with neural network-based intelligent planning. Instead of using explicit if-then rules for safety decisions, the system substitutes these with trained neural networks that can infer safe trajectories from complex patterns in data, improving reliability while reducing the computational burden of explicit rule evaluation in uncertain environments.
Data Source
AI summary
A driving route planning method applied to an electronic device is provided. The method includes acquiring an image when a vehicle is driving. Target route information is obtained by inputting the image into a target route planning model. Once a first embedding vector of the target route information is extracted, a driving route corresponding to a driving style is obtained by inputting the first embedding vector into a target driving style model.