Group Multi-Object Tracking for Dynamic Risk Map and Safe Path Planning

This paper studies the group multi-object tracking (MOT) problem in dynamic pedestrian environments, with intended application to safe navigation for autonomous vehicles. We complete a full autonomous vehicle navigation pipeline from object detection, tracking, grouping, to risk map generation and safe path planning. Our main contribution is to instantiate a group multi-object tracking algorithm, which provides the crucial grouped activity information, i. e. group position, group velocity, group size, to the risk map generator, and therewith produce a stable and robust risk map for the downstream safe path planner. Experimental results with real world data show the socially acceptable, robust and stable performance of the proposed algorithm over its individual MOT counterpart.

Authors

• Lyuyu Shen
• Hongliang Guo 0003
• Yechao Bai
• Lei Qin
• Marcelo H. Ang
• Daniela Rus

Keywords

• Measurement
• Navigation
• Heuristic algorithms
• Pipelines
• Object detection
• Path planning
• Generators
• Risk Map
• Multi-object Tracking
• Safe Path
• Safe Path Planning
• Group Size
• Dynamic Environment
• Real-world Data
• Autonomous Vehicles
• Group Velocity
• Safe Navigation
• Detection Results
• Bounding Box
• Backbone Network
• Individual Maps
• Current Frame
• Large Motion
• Previous Frame
• Tracking Results
• Unstructured Environments
• Individual Tracks
• National University Of Singapore
• Pedestrian Detection
• Dynamic Obstacles
• group-MOT
• dynamic risk map
• pedestrian environment