2021 journal article

Stochastic Gradient-Based Optimal Signal Control With Energy Consumption Bounds

IEEE Transactions on Intelligent Transportation Systems, 22(5), 3054–3067.

author keywords: Energy consumption; Optimization; Delays; Linear programming; Stochastic processes; Mathematical model; Fuels; Traffic signal optimization; fuel consumption; simultaneous perturbation stochastic approximation (SPSA); vehicle-specific power; gradient approximation; SPSA with constraints
TL;DR: A novel gradient-approximation-based solution technique offers a functional and feasible way to accommodate non-convex energy consumption bounds within a signal control optimization model to achieve maximal mobility with minimal energy consumption. (via Semantic Scholar)
UN Sustainable Development Goal Categories
7. Affordable and Clean Energy (OpenAlex)
Source: ORCID
Added: March 19, 2020

This paper develops a stochastic gradient-based optimization model for traffic signal control with bounds on network-level vehicular energy consumption. The signal control problem is formulated as a mixed-integer linear mathematical program, which incorporates inequality constraints to limit the total energy consumption in the network. The developed stochastic gradient approximation algorithm provides a near-optimal solution to the non-convex optimization problem. To account for the energy consumption constraints, a penalty function method leveraging the pseudo gradient estimation technique is developed. Empirical results from a signalized arterial street show that it is possible to achieve optimized signal settings at the desired energy consumption bound without compromising delay. Further, we report the sensitivity of the energy bounds to the mobility metrics—system delay. Our novel gradient-approximation-based solution technique offers a functional and feasible way to accommodate non-convex energy consumption bounds within a signal control optimization model to achieve maximal mobility with minimal energy consumption.