Autonomous Driving in Real-time Traffic Merging

A decentralized, computational efficient and real-time framework in SimEvents for autonomous driving in traffic merging management

This framework aims to optimize the process of Connected and Automated Vehicles (CAVs) merging at a traffic intersection while guaranteeing the state, control and safety constraints. We decompose the task of automatic merging for all the CAVs in a control zone around a merging point into same-lane safety constraints and different-lane safe merging, and implement these requirements using control barrier functions (CBFs). We consider two main objectives. First, to minimize travel time, we make the CAVs reach the road maximum speed with exponentially stabilizing control Lyapunov functions (CLF). Second, we penalize energy consumption as a cost in an optimization problem. We then decompose the merging problem into decentralized subproblems formulated as a sequence of quadratic programs (QP), which are solved in real time.

Main features:

1. All constraints (safety, control and speed) guaranteed.

2. Decentralized, computational efficient and real-time (capable to handle noise).

3. Able to solve non-linear vehicle dynamics and complex objective functions.

**Reference**: Wei Xiao, Calin Belta, and Christos G. Cassandras. Decentralized merging control in traffic networks: A control barrier function approach. In Proc. ACM/IEEE International Conference on Cyber-Physical Systems, pages 270–279, Montreal, Canada, 2019. https://doi.org/10.1145/3302509.3311054

**Joint OC and CBF approach with vehicle noise Reference**: Wei Xiao, Christos Cassandras, Calin Belta, Decentralized Merging Control in Traffic Networks with Noisy Vehicle Dynamics: a Joint Optimal Control and Barrier Function Approach. IEEE 22nd Intelligent Transportation Systems Conference (ITSC), Auckland, New Zealand, 2019

Note:
* In this version, the main and merging lanes use different CAVs generator, so two CAVs may come from these two different lanes at the same time (discretized time instant).
* If two CAVs come from two different lanes at the same time, there will be some problems because of the FIFO queue. When this happens, perturbation is enforced on the next arrival CAV (its position changes to -0.1 and the distance between i and i-1 changes to 0.1 instead of 0).

Unbalanced vehicle arrival video:
https://www.youtube.com/watch?v=Yy7k92ShCNY&t=4s

Potential to manage very heavy traffic, upper one: CBF controller, lower one: OC controller (fail to handle):
https://www.youtube.com/watch?v=yJLeS8AoU0I&t=26s

This SimEvents framework and OC controller refers to : https://www.mathworks.com/matlabcentral/fileexchange/63743-autonomous-driving-simulation-framework-for-traffic-studies