Injecting Knowledge in Data-driven Vehicle Trajectory Predictors

Abstract

Vehicle trajectory prediction tasks have been commonly tackled from two distinct perspectives: either with knowledge-driven methods or more recently with data-driven ones. On the one hand, we can explicitly implement domain-knowledge or physical priors such as anticipating that vehicles will follow the middle of the roads. While this perspective leads to feasible outputs, it has limited performance due to the difficulty to hand-craft complex interactions in urban environments. On the other hand, recent works use data-driven approaches which can learn complex interactions from the data leading to superior performance. However, generalization, \textit{i.e.}, having accurate predictions on unseen data, is an issue leading to unrealistic outputs. In this paper, we propose to learn a "Realistic Residual Block" (RRB), which effectively connects these two perspectives. Our RRB takes any off-the-shelf knowledge-driven model and finds the required residuals to add to the knowledge-aware trajectory. Our proposed method outputs realistic predictions by confining the residual range and taking into account its uncertainty. We also constrain our output with Model Predictive Control (MPC) to satisfy kinematic constraints. Using a publicly available dataset, we show that our method outperforms previous works in terms of accuracy and generalization to new scenes. We will release our code and data split here: https://github.com/vita-epfl/RRB.

arXiv commentPublished in Transportation Research: Part C

References in corpus (17)

• Adam: A Method for Stochastic Optimization
• Social Force Model for Pedestrian Dynamics
• Congested Traffic States in Empirical Observations and Microscopic Simulations
• Experimental Comparison of Open Source Visual-Inertial-Based State Estimation Algorithms in the Underwater Domain
• Informed Machine Learning -- A Taxonomy and Survey of Integrating Knowledge into Learning Systems
• INTERACTION Dataset: An INTERnational, Adversarial and Cooperative moTION Dataset in Interactive Driving Scenarios with Semantic Maps
• Physics-guided Neural Networks (PGNN): An Application in Lake Temperature Modeling
• Overcoming Limitations of Mixture Density Networks: A Sampling and Fitting Framework for Multimodal Future Prediction
• Multiple Futures Prediction
• ChauffeurNet: Learning to Drive by Imitating the Best and Synthesizing the Worst
• Neural-Symbolic Computing: An Effective Methodology for Principled Integration of Machine Learning and Reasoning
• DR-RNN: A deep residual recurrent neural network for model reduction
• Residual Policy Learning
• Social-WaGDAT: Interaction-aware Trajectory Prediction via Wasserstein Graph Double-Attention Network
• Learning in an Uncertain World: Representing Ambiguity Through Multiple Hypotheses
• Improving Deep Learning Models via Constraint-Based Domain Knowledge: a Brief Survey
• Conditional Generative Neural System for Probabilistic Trajectory Prediction

Cited by in corpus (6)

• Social Interactions for Autonomous Driving: A Review and Perspectives
• Trajectory Prediction for Autonomous Driving: Progress, Limitations, and Future Directions
• Hierarchical Motion Encoder-Decoder Network for Trajectory Forecasting
• Lifelong Vehicle Trajectory Prediction Framework Based on Generative Replay
• Addressing crash-imminent situations caused by human driven vehicle errors in a mixed traffic stream: a model-based reinforcement learning approach for CAV
• Safety-aware Motion Prediction with Unseen Vehicles for Autonomous Driving