Multi-attribute interactive visualization of three-dimensional trajectory sets

Rapidly advancing location-awareness technologies and services have collected and stored massive amounts of moving object trajectory data with attribute information that involves various degrees of spatial scales, timescales, and levels of complexity. Unfortunately, interesting behaviors regarding combinations of attributes are scarcely extracted from datasets. Further, trajectories are typically dependent on the environment of three-dimensional space, and another issue of interest to us is to preserve spatial-location visualization while guaranteeing the description of temporal information. Therefore, we developed a novel analytics tool that combines visual and interactive components to enable a dynamic visualization of three-dimensional trajectory multi-attribute behaviors. Under the context of spatiotemporal analysis, this approach integrates multiple attributes into one view to efficiently explore the attribute visualization problem of multi-attribute combination without over-plotting. To assess the feasibility of our solution, we visualized and analyzed multi-attribute information of moving object trajectories using a real mining truck dataset as a case study.