Robot trajectory prediction is an essential part of building digital twin systems and ensuring the high-performance navigation of IoT mobile robots. In the study, a novel two-stage multi-objective multi-learner model is proposed for robot trajectory prediction. Five machine learning models are adopted as base learners, including autoregressive moving average, multi-layer perceptron, Elman neural network, deep echo state network, and long short-term memory. A non-dominated sorting genetic algorithm III is applied to automatically combine these base learners, generating an accurate and robust ensemble model. The proposed model is tested on several actual robot trajectory datasets and evaluated by several metrics. Moreover, different existing optimization algorithms are also applied to compare with the proposed model. The results demonstrate that the proposed model can achieve satisfactory accuracy and robustness for different datasets. It is suitable for the accurate prediction of robot trajectory.

Autonomous quadrotor landing on moving ground platforms is essential for cooperative air–ground operations in confined warehouse environments. However, existing trajectory planners often compromise corridor clearance in narrow aisles or fail to track maneuvering platforms under dynamic intrusions. To address these challenges, a hierarchical environment-aware trajectory planning framework (HEA-Planner) is presented. At the global layer, an Environment-Aware Ant Colony Optimization (EA-ACO) algorithm is formulated to generate corridor-centered reference paths using environmental feedback. At the local layer, an Environment-Aware A* (EA-A*) replanner is designed to coordinate reactive obstacle avoidance and platform interception via spatiotemporal risk repulsion and predictive attraction fields. At the trajectory optimization layer, a Spatially-Guided MINCO (SG-MINCO) solver is introduced to preserve corridor clearance, while a Spatiotemporal Coupled MINCO (STC-MINCO) solver is derived to resolve moving-platform rendezvous through analytical duration total derivatives. Simulations across three operational states demonstrate robust collision avoidance and dynamic synchronization. Furthermore, physical flight experiments validate onboard closed-loop execution, achieving an average replanning latency of $33.6\pm 4.9\mathrm{ms}$ and a terminal touchdown error of $148.2\pm 37.3\mathrm{mm}$ across all trials (100% success rate), verifying the real-time feasibility of autonomous air–ground handover in restricted logistics workspaces.