Perception and localisation

  • Perception and localisation

This research program addresses the fundamental problem of how a robot or autonomous vehicle uses perception to create maps and calculates and tracks its location in the world. Research questions include addressing how:

  • the appearance of a place changes in relation as a function of time, season, weather, viewpoint and environment type
  • understanding context and semantics can enhance performance
  • lifelong reliability can be achieved as the world continually changes
  • the relationship to neurological structures and behavioural mechanisms are used in animal and human navigation; and how new perception technologies can be applied to this problem.

This research program builds on already considerable industry collaborations in high-value sectors including mining and defence. We look to expand into logistics, construction, space and remote operations.

We are continually building on our extensive engagement program of over 50 public and sector-based talks, workshops, media, and large-scale events each year.

Projects

A baseline dataset for performance evaluation of visual detection and classification techniques in mining environments

Active, adaptive & predictive robot localization

Adversity‐ and Adversary‐Robust Adaptive Positioning Systems with Integrity

AgBot II

Novel autonomous robotic weed control to maximise agricultural productivity

AUSMURI: Neuro-Autonomy: Neuroscience-Inspired Perception, Navigation, and Spatial Awareness for Autonomous Robots

Australian Research Council Industrial Transformation Training Centre for Joint Biomechanics

Automated Early-Detection of the Invasive Grass African Lovegrass

Autonomous Mission Planning, Navigation and Geological Feature Recognition using UAVs (Drones)

Autonomous UAV decision making under environment and target detection uncertainty

[COMPLETED] ARC Future Fellowship: Superhuman place recognition

Automation-enabling positioning for underground mining

[COMPLETED] How automated vehicles will interact with road infrastructure

Contextual Hazard Detection

Continuum robots

Effect of lighting on visual odometry performance in underground mines

[COMPLETED] HD maps for automated driving

Cameras and sensors

Mini Autonomous Vehicles

Neuromorphic computing

Re-Evolving Nature’s Best Positioning Systems for People and Their Machines

Rheinmetall Defence Australia: Advanced Terrain Detection (ATD)

Robotic knee arthroscopy

Spike-based Visual Place Recognition using Intel’s Loihi

UAV Navigation using semantic cues

US AFOSR / AOARD: An infinitely scalable learning and recognition network

Visual Place Recognition for Robotics in Extreme Environments

PhD Topics

Team

Program Lead: Professor Michael Milford

Deputy: Professor Felipe Gonzalez

Key Research Fellows: Dr Tobias Fischer