NeuralMOVES: A lightweight and microscopic vehicle emission estimation model based on reverse engineering and surrogate learning

2025-02-06

AI Summary

To address the high computational complexity of the MOVES model, which hinders real-time, vehicle-level emission estimation, this work proposes a lightweight neural network surrogate model. We introduce the first industrial-software reverse-engineering framework tailored to traffic scenarios, enabling knowledge distillation and structural compression of MOVES—reducing its size to 2.4 MB (>99% reduction). The resulting multilayer perceptron (MLP) surrogate integrates trajectory-, environment-, and vehicle-level features. Evaluated on over two million real-world scenarios, it achieves a mean absolute percentage error of 6.013% and enables millisecond-scale CO₂ emission inference per vehicle, with full edge-deployment capability. This work breaks the longstanding trade-off between accuracy and efficiency in microscopic emission modeling, establishing a practical, real-time paradigm for emission-reduction strategy optimization.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Planning, Routing, and Scheduling: Model-Based Reasoning
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

The transportation sector significantly contributes to greenhouse gas emissions, necessitating accurate emission models to guide mitigation strategies. Despite its field validation and certification, the industry-standard Motor Vehicle Emission Simulator (MOVES) faces challenges related to complexity in usage, high computational demands, and its unsuitability for microscopic real-time applications. To address these limitations, we present NeuralMOVES, a comprehensive suite of high-performance, lightweight surrogate models for vehicle CO2 emissions. Developed based on reverse engineering and Neural Networks, NeuralMOVES achieves a remarkable 6.013% Mean Average Percentage Error relative to MOVES across extensive tests spanning over two million scenarios with diverse trajectories and the factors regarding environments and vehicles. NeuralMOVES is only 2.4 MB, largely condensing the original MOVES and the reverse engineered MOVES into a compact representation, while maintaining high accuracy. Therefore, NeuralMOVES significantly enhances accessibility while maintaining the accuracy of MOVES, simplifying CO2 evaluation for transportation analyses and enabling real-time, microscopic applications across diverse scenarios without reliance on complex software or extensive computational resources. Moreover, this paper provides, for the first time, a framework for reverse engineering industrial-grade software tailored specifically to transportation scenarios, going beyond MOVES. The surrogate models are available at https://github.com/edgar-rs/neuralMOVES.

Problem

• Develops lightweight vehicle CO2 emission model
• Reduces computational complexity of MOVES
• Enables real-time microscopic emission analysis

Innovation

• Lightweight surrogate models for CO2 emissions
• Reverse engineering industrial-grade transportation software
• Neural Networks achieving high accuracy