Recursive Least Squares Based Refinement Network for the Rollout Trajectory Prediction Methods
This work addresses trajectory prediction challenges for intelligent vehicles, but it appears incremental as it refines existing rollout methods with a plug-in module.
The paper tackles the problem of accumulative error and weak adaptability in rollout trajectory prediction for intelligent vehicles by proposing a parametric-learning recursive least squares (RLS) estimation based on deep neural network, with experiments on the NGSIM dataset showing effective improvement.
Trajectory prediction plays a pivotal role in the field of intelligent vehicles. It currently suffers from several challenges,e.g., accumulative error in rollout process and weak adaptability in various scenarios. This paper proposes a parametric-learning recursive least squares (RLS) estimation based on deep neural network for trajectory prediction. We design a flexible plug-in module which can be readily implanted into rollout approaches. Goal points are proposed to capture the long-term prediction stability from the global perspective. We carried experiments out on the NGSIM dataset. The promising results indicate that our method could improve rollout trajectory prediction methods effectively.