Complementary Learning System Empowers Online Continual Learning of Vehicle Motion Forecasting in Smart Cities

2025-08-27

AI Summary

To address catastrophic forgetting in continual learning of deep neural networks for intelligent city vehicle motion prediction, this paper proposes the Dual-LS framework—the first to incorporate the human brain’s complementary learning mechanism into this task. It introduces a task-agnostic, online-updatable dual-memory replay system: a long-term memory preserves generalized knowledge, while a short-term memory rapidly adapts to evolving traffic patterns; their synergy enables efficient experience retrieval and balanced knowledge retention. Crucially, the framework avoids storing or replaying raw trajectory data, substantially reducing computational overhead. Evaluated on a large-scale multi-country dataset comprising 772,000 vehicles and 11,000 km of real-world trajectories, Dual-LS reduces catastrophic forgetting by 74.31% and cuts computational resource consumption by 94.02% compared to baseline methods. Moreover, it significantly enhances long-horizon prediction stability and environmental adaptability.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Learning for Planning and Scheduling

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Artificial intelligence underpins most smart city services, yet deep neural network (DNN) that forecasts vehicle motion still struggle with catastrophic forgetting, the loss of earlier knowledge when models are updated. Conventional fixes enlarge the training set or replay past data, but these strategies incur high data collection costs, sample inefficiently and fail to balance long- and short-term experience, leaving them short of human-like continual learning. Here we introduce Dual-LS, a task-free, online continual learning paradigm for DNN-based motion forecasting that is inspired by the complementary learning system of the human brain. Dual-LS pairs two synergistic memory rehearsal replay mechanisms to accelerate experience retrieval while dynamically coordinating long-term and short-term knowledge representations. Tests on naturalistic data spanning three countries, over 772,000 vehicles and cumulative testing mileage of 11,187 km show that Dual-LS mitigates catastrophic forgetting by up to 74.31% and reduces computational resource demand by up to 94.02%, markedly boosting predictive stability in vehicle motion forecasting without inflating data requirements. Meanwhile, it endows DNN-based vehicle motion forecasting with computation efficient and human-like continual learning adaptability fit for smart cities.

Problem

• Mitigates catastrophic forgetting in neural networks
• Enables online continual learning for motion forecasting
• Balances long-term and short-term knowledge representation

Innovation

• Dual-LS paradigm with complementary memory systems
• Synergistic rehearsal mechanisms for experience retrieval
• Dynamic coordination of long-term and short-term knowledge