Search papers, labs, and topics across Lattice.
This paper introduces DIFF-IPPO, a novel pipeline that combines an open-vocabulary belief map generator with a diffusion-based planner to enhance global trajectory generation for robots in exploration and object search tasks. By leveraging complex, multimodal belief maps, the method effectively concentrates sensor coverage in high-belief regions, resulting in impressive normalized detection scores ranging from 81.49% to 86.55%. In a simulated search-and-rescue scenario, a team of five drones using this approach achieved first detections in just 3.5 minutes, showcasing its practical effectiveness in urgent applications.
Achieving first detections in a simulated search-and-rescue mission in just 3.5 minutes, DIFF-IPPO redefines the efficiency of robotic exploration.
Exploration and object search require robots to perceive their environment, identify regions of interest, and plan trajectories that improve target-detection likelihood or maximize information gain. Many IPP methods, especially in continuous environmental monitoring, rely on Gaussian-process belief models, while object-search settings often produce complex, multimodal belief maps from semantic or open-vocabulary perception. Global trajectory generation directly conditioned on such non-Gaussian belief maps remains comparatively underexplored. Although diffusion-based planners offer strong capabilities for modeling such distributions, their use in informative path planning remains limited. In this work, we propose DIFF-IPPO, a pipeline that integrates an open-vocabulary belief map generator with a diffusion-based planner for global trajectory generation over belief maps. The method generates trajectories that concentrate sensor coverage over high-belief regions, achieving normalized detection scores between 81.49% and 86.55% across different dataset scenarios. We validate the system in a simulated search-and-rescue scenario where the planner searches candidate building regions to locate a burning building. In this setting, a team of five drones using batched belief-map-conditioned trajectory generation achieves first detections in 3.5 minutes.
