DynORecon: Dynamic Object Reconstruction for Navigation

This paper presents DynORecon, a Dynamic Object Reconstruction system that leverages the information provided by Dynamic SLAM to simultaneously generate a volumetric map of observed moving entities while estimating free space to support navigation. By capitalising on the motion estimations provided by Dynamic SLAM, DynORecon continuously refines the representation of dynamic objects to eliminate residual artefacts from past observations and incrementally reconstructs each object, seamlessly integrating new observations to capture previously unseen structures. Our system is highly efficient (~20 FPS) and produces accurate (~10 cm) object reconstructions using simulated and real-world outdoor datasets.

Authors

• Yiduo Wang 0001
• Jesse Morris
• Lan Wu
• Teresa A. Vidal-Calleja
• Viorela Ila

Keywords

• Simultaneous localization and mapping
• Accuracy
• Navigation
• Motion estimation
• Dynamics
• Robotics and automation
• Dynamic Objects
• Object Reconstruction
• Dynamic Object Reconstruction
• System Dynamics
• Free Space
• Residual Artifacts
• Reconstruction System
• Reference Frame
• Dynamic Environment
• Rigid Body
• Point Cloud
• Global Map
• 3D Point
• Object Motion
• Obstacle Avoidance
• Static Function
• Navigation Task
• Dynamic Point
• Dense Reconstruction
• Efficient Update
• Occupied Space
• World Frame
• Object Pose
• Large-scale Environments
• Signed Distance Function
• Static Background
• Object In Frame
• Ground Truth Map