Resource-aware Online Parameter Adaptation for Computationally-constrained Visual-Inertial Odometry @autonomousrobotslab

Uploaded November 2020 | Updated September 2026, 3 weeks ago

In this work, a computational resources-aware parameter adaptation method for visual-inertial navigation systems is proposed with the goal of enabling the improved deployment of such algorithms on computationally constrained systems. Such a capacity can prove critical when employed on ultra-lightweight systems or alongside mission critical computationally expensive processes. To achieve this objective, the algorithm proposes selected changes in the vision front-end and optimization back-end of visual-inertial odometry algorithms, both prior to execution and in real time based on an online profiling of available resources. The method also utilizes information from the motion dynamics experienced by the system to manipulate parameters online. The general policy is demonstrated on three established algorithms, namely S-MSCKF, VINS-Mono and OKVIS and has been verified experimentally on the EuRoC dataset. The proposed approach achieved comparable performance at a fraction of the original computational cost.

$Resource-aware Online Parameter Adaptation for Computationally-constrained Visual-Inertial Odometry In this work, a computational resources-aware parameter adaptation method for visual-inertial navigation systems is proposed with the goal of enabling the improved deployment of such algorithms on computationally constrained systems. Such a capacity can prove critical when employed on ultra-lightweight systems or alongside mission critical computationally expensive processes. To achieve this objective, the algorithm proposes selected changes in the vision front-end and optimization back-end of visual-inertial odometry algorithms, both prior to execution and in real time based on an online profiling of available resources. The method also utilizes information from the motion dynamics experienced by the system to manipulate parameters online. The general policy is demonstrated on three established algorithms, namely S-MSCKF, VINS-Mono and OKVIS and has been verified experimentally on the EuRoC dataset. The proposed approach achieved comparable performance at a fraction of the original computational cost.$

Autonomous Exploration inside Building Corridors

A Multi-Modal Mapping Unit for Autonomous Robotic Exploration in Visually-degraded Environments

Model Predictive Attitude Control of a Jumping-and-Flying Quadruped for Planetary Exploration

Exploration and Mapping in Visually-degraded Environments using Aerial Robots - Preliminary Results

OmniPlanner: Universal Exploration and Inspection Path Planning across Robot Morphologies

DARPA SubT Urban Circuit: Autonomous Exploration in the Satsop Abandoned Power Plant

Model Predictive Control for Micro Aerial Vehicles: A Survey

ICRA2024 Talk: Reinforcement Learning for Collision-free Flight Exploiting Deep Collision Encoding

Towards Quadrupedal Jumping and Walking for Dynamic Locomotion using Reinforcement Learning

Autonomous Teamed Exploration of Subterranean Environments - Simulation study

Autonomous Distributed Radiation Field Characterization and Informative Planning: Experiment #2