Arrow Research search
Back to IROS

IROS 2018

Reactive Collision Avoidance Using Real-Time Local Gaussian Mixture Model Maps

Conference Paper Accepted Paper Artificial Intelligence ยท Robotics

Abstract

In unknown, cluttered environments, robots require online real-time mapping and collision checking in order to navigate robustly. Discrete map representations are inefficient for collision checking as they are expensive in terms of memory and computation. This paper takes a probabilistic approach to local mapping by representing the environment as a Gaussian Mixture Model (GMM) and leverages its geometric properties to enable efficient collision checking given a time-parameterized trajectory. In contrast to current discretization-based methods, a GMM preserves geometric coverage of the environment without losing representation accuracy with varying map resolutions. We introduce a novel GMM local mapping algorithm that can be used with a single depth camera processed on a single CPU, and provide algorithms for collision avoidance given arbitrary trajectory representations. Finally, we provide experimentation results demonstrating safety, efficiency, and data coverage for real-time collision avoidance with a quadrotor navigating in a cluttered environment.

Authors

Keywords

  • Trajectory
  • Collision avoidance
  • Robot sensing systems
  • Real-time systems
  • Current measurement
  • Gaussian mixture model
  • Local Map
  • Real-time Mapping
  • Reactive Collision Avoidance
  • Depth Camera
  • Map Representation
  • Collision Detection
  • Single CPU
  • Discretion
  • Computational Complexity
  • Angular Velocity
  • Linear Approximation
  • Depth Images
  • Sensor Measurements
  • Configuration Space
  • Linear Velocity
  • Current Sensor
  • Current Frame
  • Mean Component
  • Motion Primitives
  • Gaussian Components
  • Treemap
  • Local Trajectory
  • Sensor Observations
  • Trajectory Curve
  • Probable Point
  • Current Pose
  • Input Space
  • Piecewise Affine

Context

Venue
IEEE/RSJ International Conference on Intelligent Robots and Systems
Archive span
1988-2025
Indexed papers
26578
Paper id
1107891086335714384
v2026.09.13