In this project, we present Distributed DeeP-LCC (Data-EnablEd Predictive Leading Cruise Control) algorithm for CAV cooperation in large-scale mixed traffic flow.
DeeP-LCC is a centralized data-driven predictive control strategy for CAVs in mixed traffic, where human-driven vehicles (HDVs) also exist. It collects the measurable data from the entire mixed traffic system, and relies on Willems' Fundamental Lemma for behavior representation and predictive control. See Project DeeP-LCC for details.
Cooperative DeeP-LCC naturally partitions the mixed traffic system into multiple CF-LCC (Car-Following LCC) subsystems, with one leading CAV and multiple HDVs following behind (if they exist). Each CAV directly utilizes measurable traffic data from its own CF-LCC subsystem to design safe and optimal control behaviors. The interaction between neighbouring subsystems is formulated as a coupling constraint.
The optimization problem is as follows.
A tailored ADMM based distributed implementation algorithm (distributed DeeP-LCC) is designed to solve the cooperative DeeP-LCC formulation.
The benefits of our algorithm include:
- Computation efficiency
- Communication efficiency
- Local data privacy
To contact us about DeeP-LCC, email Jiawei Wang.