Towards Scalable & Efficient Interaction-Aware Planning in Autonomous Vehicles using Knowledge Distillation
Preprint
Abstract
Real-world driving involves intricate interactions among vehicles navigating through dense traffic scenarios. Recent research focuses on enhancing the interaction awareness of autonomous vehicles to leverage these interactions in decision-making. These interaction-aware planners rely on neural-network-based prediction models to capture inter-vehicle interactions, aiming to integrate these predictions with traditional control techniques such as Model Predictive Control. However, this integration of deep learning-based models with traditional control paradigms often results in computationally demanding optimization problems, relying on heuristic methods. This study introduces a principled and efficient method for combining deep learning with constrained optimization, employing knowledge distillation to train smaller and more efficient networks, thereby mitigating complexity. We demonstrate that these refined networks maintain the problem-solving efficacy of larger models while significantly accelerating optimization. Specifically, in the domain of interaction-aware trajectory planning for autonomous vehicles, we illustrate that training a smaller prediction network using knowledge distillation speeds up optimization without sacrificing accuracy.Keywords
All Related Versions
This publication has 9 references indexed in Scilit:
- Interaction-Aware Motion Planning for Autonomous Vehicles With Multi-Modal Obstacle Uncertainty PredictionsIEEE Transactions on Intelligent Vehicles, 2023
- Towards Physically Adversarial Intelligent Networks (PAINs) for Safer Self-DrivingIEEE Control Systems Letters, 2022
- A Cooperation-Aware Lane Change Method for Automated VehiclesIEEE Transactions on Intelligent Transportation Systems, 2022
- Lane-Change in Dense Traffic With Model Predictive Control and Neural NetworksIEEE Transactions on Control Systems Technology, 2022
- A Review of End-to-End Autonomous Driving in Urban EnvironmentsIEEE Access, 2022
- Graph-Based Spatial-Temporal Convolutional Network for Vehicle Trajectory Prediction in Autonomous DrivingIEEE Transactions on Intelligent Transportation Systems, 2022
- Distributed Optimization and Statistical Learning via the Alternating Direction Method of MultipliersFoundations and Trends in Machine Learning, 2011
- Neural network based model predictive control for a steel pickling processJournal of Process Control, 2009
- Large-scale nonlinear programming using IPOPT: An integrating framework for enterprise-wide dynamic optimizationComputers & Chemical Engineering, 2009