학술논문

KI-PMF: Knowledge Integrated Plausible Motion Forecasting
Document Type
Conference
Source
2024 IEEE Intelligent Vehicles Symposium (IV) Intelligent Vehicles Symposium (IV), 2024 IEEE. :176-183 Jun, 2024
Subject
Communication, Networking and Broadcast Technologies
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Transportation
Knowledge engineering
Accuracy
Roads
Kinematics
Trajectory
Topology
Safety
Motion Forecasting and Planning
Knowledge Integration
Language
ISSN
2642-7214
Abstract
The accurate prediction of surrounding traffic actors’ movements is vital for the large-scale safe deployment of autonomous vehicles. Existing motion forecasting methods primarily aim to minimize prediction error by optimizing a loss function, which can sometimes lead to physically infeasible predictions or states that violate external constraints. This paper proposes a method that integrates explicit knowledge priors, allowing a network to forecast future trajectories that comply with both the vehicle’s kinematic constraints and the driving environment’s geometry. This is achieved by introducing a non-parametric pruning layer, and learnable attention layers to incorporate the defined knowledge priors. The proposed method aims to ensure reachability guarantees for traffic actors in both complex and dynamic situations. By conditioning the network to adhere to physical laws, we can achieve accurate and safe predictions, which are crucial for maintaining the safety and efficiency of autonomous vehicles in real-world settings.