Predicting Human Trajectories by Learning and Matching Patterns
This addresses the need for robots to navigate safely and intuitively in human-shared spaces, though it appears incremental as it builds on existing CNN approaches with a focus on explainability.
The paper tackles the problem of predicting human trajectories for robot navigation by proposing a Convolutional Neural Network-based method called Social-PEC, which learns and matches patterns in sequential data, achieving performance comparable to state-of-the-art and outperforming in some cases.
As more and more robots are envisioned to cooperate with humans sharing the same space, it is desired for robots to be able to predict others' trajectories to navigate in a safe and self-explanatory way. We propose a Convolutional Neural Network-based approach to learn, detect, and extract patterns in sequential trajectory data, known here as Social Pattern Extraction Convolution (Social-PEC). A set of experiments carried out on the human trajectory prediction problem shows that our model performs comparably to the state of the art and outperforms in some cases. More importantly, the proposed approach unveils the obscurity in the previous use of a pooling layer, presenting a way to intuitively explain the decision-making process.