Obstacle‐transformer: A trajectory prediction network based on surrounding trajectories

Abstract

Recurrent Neural Network, Long Short-Term Memory, and Transformer have made great progress in predicting the trajectories of moving objects. Although the trajectory element with the surrounding scene features has been merged to improve performance, there still exist some problems to be solved. One is that the time series processing models will increase the inference time with the increase of the number of prediction sequences. Another problem is that the features cannot be extracted from the scene's image and point cloud in some situations. Therefore, an Obstacle-Transformer is proposed to predict trajectory in a constant inference time. An 'obstacle' is designed by the surrounding trajectory rather than images or point clouds, making Obstacle-Transformer more applicable in a wider range of scenarios. Experiments are conducted on ETH and UCY datasets to verify the performance of our model.

A Review of Trajectory Prediction Methods for the Vulnerable Road User

“…Apart from methods that use the original Transformer architecture, this chapter will also introduce methods that use an attention mechanism to predict trajectories. The Transformer architecture can be used without any changes to predict trajectories based on observed positions [17,58,59,[74][75][76].…”

“…Zhang et al [58] applied two Transformers in parallel, one to infer obstacle positions from observed trajectories and one for predicting the x and y positions separately.…”

“…Local frames of reference require transforming agent positions into every agent's frame of reference, which benefits the pose invariance of the model but scales linearly with the number of agents [17]. Zhang et al [58] employed a third approach to parse the 2D positions, supplying the method with the trajectory of each agent as a sequence of relative positions, or position changes.…”