Dynamic Platoon Formation in Mixed Autonomous Vehicle Traffic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle platooning methods fail to effectively account for mixed traffic flows involving human-driven vehicles (HDVs) and connected autonomous vehicles (CAVs), leading to inefficiencies in platoon formation and energy consumption.
Innovation Solution
A dynamic platoon formation method that involves sending and receiving communication signals among autonomous vehicles, obtaining road and vehicle information, and using a decision tree process to select the optimal lane for platooning based on indices such as time to platoon (TTP) and distance to platoon (DTP), considering the actions and positions of both HDVs and CAVs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If autonomous vehicles form a platoon in mixed traffic flow with human-driven vehicles, then energy consumption is reduced due to decreased wind resistance, but the complexity of coordinating vehicle actions increases because human-driven vehicles are not under control
Solution Approach 1:
The patent segments the traffic flow into different lanes and processes platoon formation for each lane separately. The decision tree is constructed by evaluating each candidate lane independently, dividing the complex coordination problem into manageable sub-problems that can be solved sequentially rather than attempting to coordinate all vehicles simultaneously across all lanes.
Solution Approach 2:
The patent performs preliminary actions by pre-calculating decision trees for each candidate lane before final platoon formation. The system evaluates TTP and DTP metrics for each lane in advance, identifies optimal decision sequences beforehand, and only then executes the platoon formation maneuver, reducing real-time coordination complexity.
2Adaptability or versatility
If the system evaluates multiple candidate lanes for platoon formation, then the applicability to diverse traffic scenarios is improved, but the computation time and complexity increase
Solution Approach 1:
The patent divides the multi-lane evaluation problem into separate decision tree constructions for each candidate lane. Instead of evaluating all lanes simultaneously in a single complex computation, the system processes each lane independently, constructing individual decision trees that can be evaluated and compared afterward.
Solution Approach 2:
The patent implements dynamic lane selection by constructing decision trees that adapt to current traffic conditions in each lane. The system dynamically evaluates TTP and DTP metrics based on real-time vehicle positions and speeds, allowing the optimal lane to be determined adaptively rather than using a fixed evaluation framework.
3Reliability
If the platoon formation method considers actions of human-driven vehicles, then the reliability of platoon formation is improved, but the difficulty of detecting and measuring traffic flow state increases
Solution Approach 1:
The patent incorporates feedback mechanisms by continuously monitoring the traffic flow state and updating the decision tree construction accordingly. The system detects the positions and speeds of both CAVs and HDVs, uses this information to evaluate candidate lanes, and adjusts the platoon formation strategy based on the detected state, creating a closed-loop control system that improves reliability.
Solution Approach 2:
The patent uses the decision tree as an intermediary structure that mediates between the complex traffic flow state detection and the platoon formation decision. The decision tree transforms raw detection data about HDV and CAV positions into structured evaluation metrics (TTP, DTP) that guide reliable platoon formation without requiring direct complex coordination.
4Productivity
If the system selects the optimal decision sequence based on minimum TTP or DTP, then the productivity of platoon formation is improved, but the loss of time for computing and selecting the optimal sequence increases
Solution Approach 1:
The patent performs preliminary computation by constructing complete decision trees for all candidate lanes before selecting the optimal sequence. The TTP and DTP metrics are calculated in advance for each lane's decision tree, allowing the system to quickly select the optimal sequence without performing complex real-time optimization during actual platoon formation.
Solution Approach 2:
The patent segments the optimization problem into separate decision tree constructions for each lane, with each tree evaluating a specific lane's potential. This division allows parallel or sequential processing of simpler sub-problems rather than attempting to optimize all lanes simultaneously, reducing the overall computation time while maintaining productivity.
Data Source
AI summary
A dynamic platoon formation method under a mixed autonomous vehicles flow is provided. The method implements dynamic platooning by taking into account a fact that a traffic flow is a mixture of HDVs and CAVs. The dynamic platoon formation method includes: selecting lanes as candidate lanes in turn; constructing a decision tree from a current moment to a moment of platoon formation according to the following process: constructing a decision space for each CAV, generating a compatible decision set, selecting and executing a compatible decision, and updating location and speed information of all vehicles; and selecting, according to a predetermined index (including TTP and DTP), an optimal decision sequence as a decision sequence corresponding to the candidate lane.