Autonomous vehicles are cyber-physical systems that combine embedded computing and deep learning with physical systems to perceive the world, predict future states, and safely control the vehicle through changing environments. The ability of an autonomous vehicle to accurately predict the motion of other road users across a wide range of diverse scenarios is critical for both motion planning and safety. However, existing motion prediction methods do not explicitly model contextual information about the environment, which can cause significant variations in performance across diverse driving scenarios. To address this limitation, we propose
Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher.
Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?
Some links on this page may take you to non-federal websites. Their policies may differ from this site.
-
CASTNet : a dynamic, context-aware approach for motion prediction that (i) identifies the current driving context using a spatio-temporal model, (ii) adapts an ensemble of motion prediction models to fit the current context, and (iii) applies novel trajectory fusion methods to combine predictions output by the ensemble. This approach enables CASTNet to improve robustness by minimizing motion prediction error across diverse driving scenarios. CASTNet is highly modular and can be used with various existing image processing backbones and motion predictors. We demonstrate how CASTNet can improve both CNN-based and graph-learning-based motion prediction approaches and conduct ablation studies on the performance, latency, and model size for various ensemble architecture choices. In addition, we propose and evaluate several attention-based spatio-temporal models for context identification and ensemble selection. We also propose a modular trajectory fusion algorithm that effectively filters, clusters, and fuses the predicted trajectories output by the ensemble. On the nuScenes dataset, our approach demonstrates more robust and consistent performance across diverse, real-world driving contexts than state-of-the-art techniques.Free, publicly-accessible full text available April 30, 2025 -
Integration of distributed renewable energy sources (D- RES) has been introduced as a viable solution to offer cheap and clean energy to customers in decentralized power system. D- RES can offer local generation to flexible customers based on their servicing deadline and constraints, benefiting both D- RES owners and customers in terms of providing economic revenue and reducing the cost of supplied energy. In this context, this paper proposes a dynamic matching framework using model predictive control (MPC) to enable local energy sharing in power system operation. The proposed matching framework matches flexible customers with D- RES to maximize social welfare in the matching market, while meeting the customers' servicing constraints prior to their deadline. Simulations are conducted on a test power system using multiple matching algorithms across different load and generation scenarios and the results highlighted the efficiency of proposed framework in matching flexible customers with the appropriate supply sources to maximize social welfare in the matching market.more » « less