Adaptive network improves human motion prediction around robots
Adaptive Interaction Network for Human Motion Prediction During Human-Robot Collaboration
Computer Vision and Pattern Recognition
Summary
Predicting how people will move near robots helps make interactions safer and more efficient. The authors designed a model that pays attention to how robots affect human movement and predicts future human poses accordingly. Their approach uses two parts working together—one predicts human motion while the other predicts robot motion—and a clever method to combine information locally and globally. They show their method works better than others on multiple test sets.
What this means in practice
- •For industrial robotics teams: Improve robots' ability to predict co-workers' movements in shared workspaces, enhancing safety and efficiency.
- •For robotic system integrators: Refine robot motion planning by incorporating accurate predictions of human actions influenced by robot presence.
Authors
Mengyuan Liu, Yangting Lin, Qiongjie Cui
Abstract
Human motion prediction during human-robot collaboration is critical for achieving safe and efficient interactions in shared environments. Unlike previous studies that have primarily focused on humans and objects, we focus on the robot-aware human motion prediction task, which explicitly models the influence of robots on human motion. This task presents new challenges, such as capturing human-robot heterogeneity and modeling complex spatial interactions. To address these issues, we develop an adaptive interaction network (AINet) model that consists of two branches: a primary branch that predicts future human motion and an auxiliary branch that estimates robot trajectories. The two branches are jointly optimized and coupled via a Local-Global Spatial Interaction (LGSI) Module, which effectively captures fine-grained and global contextual dependencies between human and robot motion sequences. Then, we introduce an Adaptive Weighted Aggregation (AWA) Module to dynamically fuse motion features using context-dependent weights, thereby increasing adaptability across diverse scenarios. Furthermore, we adopt a coarse-to-fine prediction strategy, in which a coarse pseudo-trajectory is first predicted, followed by the refinement of detailed human poses based on that trajectory. Extensive experiments on three challenging datasets reveal that our proposed method achieves superior performance, validating its effectiveness in human-robot collaboration scenarios. Our code is available at: https://github.com/LyTingHub/AINet.