Autonomous driving for vehicular networks with nonlinear dynamics

Abstract

In this paper, we introduce cooperative autonomous driving algorithms for vehicular networks with nonlinear mobile robot dynamics in urban environments that take human safety into account and are capable of performing vehicle-to-vehicle (V2V) and vehicle-to-pedestrian (V2P) collision avoidance. We argue that “flocks” are multi-agent models of vehicular traffic on roads and propose novel autonomous driving architectures and algorithms for cyber-physical vehicles capable of performing autonomous driving tasks such as lane-driving, lane-changing, braking, passing, and making turns. Our proposed autonomous driving algorithms are inspired by Olfati-Saber's flocking theory. Though, there are notable differences between autonomous driving on urban roads and flocking behavior - flocks have a single desired destination whereas most drivers on road do not share the same destination. We refer to this collective behavior (driving) as “multi-objective flocking.” The self-driving vehicles in our framework turn out to be hybrid systems with a finite number of discrete states that are related to the driving modes of vehicles. Complex driving maneuvers can be performed using a sequence of mode switchings. We use near-identity nonlinear transformations to extend the application of particle-based autonomous driving algorithms to multi-robot networks with nonlinear dynamics. The derivation of the mode switching conditions that preserve safety is non-trivial and an important part of the design of autonomous driving algorithms. We present several examples of driving tasks that can be effectively performed using our proposed driving algorithms.