A Delay Compensation Framework Based on Eye-Movement for Teleoperated Ground Vehicles
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This paper addresses the performance degradation of teleoperated ground vehicles caused by communication delays, which lead to asynchrony between operator commands and vehicle responses. When delays exceed 200 ms, human operators struggle to adapt, resulting in oversteering, oscillations, and increased cognitive workload. To mitigate this, the authors propose an eye-movement-based predicted trajectory guidance control (ePTGC) framework. This approach extracts human intention from eye-movement data to generate guidance trajectories, effectively removing the operator from the direct control loop and reducing sensitivity to delay. The ePTGC framework consists of trajectory prediction and tracking modules. The prediction module utilizes a parallel-structured model that integrates historical vehicle states, control commands, and visual Area of Interest (AOI) derived from eye-tracking. It employs LSTM networks for motion encoding and ResNet for context encoding using 3D LiDAR point clouds converted into binary bird’s-eye-view images. A multi-head attention mechanism fuses these features to predict driving intentions (straight, left, or right turn) and corresponding multimodal trajectories. The trajectory with the highest probability is selected to guide the vehicle via a Stanley controller on the vehicle side. The study evaluates the framework through a human-in-the-loop simulation platform using Carla and Trucksim, with delays simulated via a FIFO pipeline. Experiments involved participants completing driving tasks across five delay levels (200 ms to 1000 ms). Performance was measured using deviation to centerline (D2C), task completion time (TCT), and steering effort (SE). Results from repeated measures ANOVA indicate that the ePTGC method significantly improves maneuverability and reduces cognitive burden at large delay levels (>200 ms) compared to traditional methods. The proposed model also demonstrated superior trajectory prediction accuracy, particularly in reducing large final displacement errors, by leveraging the synergistic effect of eye-movement and contextual features. The significance of this work lies in its model-free approach to delay compensation, which avoids reliance on accurate vehicle dynamics models. By utilizing early intention cues from eye-movement, the framework provides timely and accurate trajectory predictions, enhancing system stability and operator efficiency. This method offers a robust solution for teleoperation systems operating under significant communication latency, improving both safety and operational performance in military and civilian applications.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | OpenAlex-citations | — | — | 1 | 2026-06-20 |
| archive | success | semantic_scholar | — | — | 6 | 2026-06-26 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-20 |
| chunk | success | chunk | — | — | 1 | 2026-06-20 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-20 |
| promote | success | — | — | — | 1 | 2026-06-20 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-20 |
| verify | success | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Methodological Resource: tool software