Hierarchical Multi-Channel Tip-Over Prediction via Isolated Temporal Models for Planetary Rover Safety
Hierarchical Multi-Channel Tip-Over Prediction via Isolated Temporal Models for Planetary Rover Safety
This research introduces a hierarchical multi-channel tip-over prediction framework designed for planetary rovers operating in unstructured environments. By integrating a multi-horizon Temporal Convolutional Network (TCN) to learn instability patterns from inertial measurements with a Gated Recurrent Unit (GRU)-based forward dynamics model, the system detects both gradual evolution and abrupt disturbances. Deployed as an Orocos Real-Time Toolkit component within the Rock robotics framework at 100 Hz, experiments on IMU samples across heterogeneous terrains demonstrate 97.58% event prediction coverage with a 1.82s median lead time, outperforming individual channels. The results validate that isolated temporal models combined through hierarchical decision fusion provide a robust solution for real-time rover safety.
The study presents a hierarchical multi-channel tip-over prediction framework for planetary rovers, combining inertial measurements with forward dynamics models to achieve 97.58% event prediction coverage and a 1.82s median lead time on IMU samples from heterogeneous terrains.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
seconds median lead time
1.82
View original evidence
terrains demonstrate 97.58% event prediction coverage with a 1.82s median lead time, outperforming individual prediction channels. The results validate that isolated temporal models combined through hierarchical decision fusion provide a robust solution for real-time rover safety in uncertain terrain.
Open source S5frequency of operation
100
View original evidence
rated through decision-level fusion and priority-based arbitration to generate graded safety actions, including early warning and emergency braking. The proposed system is deployed as an Orocos Real-Time Toolkit (RTT)-based safety component within the Rock robotics framework, operating at 100 Hz on a planetary rover software stack. Experiments on Inertial Measurement Unit (IMU) samples collected across heterogeneous
Open source S4
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
terrains demonstrate 97.58% event prediction coverage with a 1.82s median lead time, outperforming individual prediction channels. The results validate that isolated temporal models combined through hierarchical decision fusion provide a robust solution for real-time rover safety in uncertain terrain.
Open source S5
rated through decision-level fusion and priority-based arbitration to generate graded safety actions, including early warning and emergency braking. The proposed system is deployed as an Orocos Real-Time Toolkit (RTT)-based safety component within the Rock robotics framework, operating at 100 Hz on a planetary rover software stack. Experiments on Inertial Measurement Unit (IMU) samples collected across heterogeneous
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org