Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2401.03944

Diegetic Graphical User Interfaces and Intuitive Control of Assistive Robots via Eye-gaze

Emanuel Nunez Sardinha·Marcela Munera·Nancy Zook·David Western·Virginia Ruiz Garate
View original source ↗Read source document ↗Export metadata ↓

Abstract

Individuals with tetraplegia and similar forms of paralysis suffer physically and emotionally due to a lack of autonomy. To help regain part of this autonomy, assistive robotic arms have been shown to increase living independence. However, users with paralysis pose unique challenging conditions for the control of these devices. In this article, we present the use of Diegetic Graphical User Interfaces, a novel, intuitive, and computationally inexpensive approach for gaze-controlled interfaces applied to robots. By using symbols paired with fiducial markers, interactive buttons can be defined in the real world which the user can trigger via gaze, and which can be embedded easily into the environment. We apply this system to pilot a 3-degree-of-freedom robotic arm for precision pick-and-place tasks. The interface is placed directly on the robot to allow intuitive and direct interaction, eliminating the need for context-switching between external screens, menus, and the robot. After calibration and a brief habituation period, twenty-one participants from multiple backgrounds, ages and eye-sight conditions completed the Yale-CMU-Berkeley (YCB) Block Pick and Place Protocol to benchmark the system, achieving a mean score of 13.71 out of the maximum 16.00 points. Good usability and user experience were reported (System Usability Score of 75.36) while achieving a low task workload measure (NASA-TLX of 44.76). Results show that users can employ multiple interface elements to perform actions with minimal practice and with a small cognitive load. To our knowledge, this is the first easily reconfigurable screenless system that enables robot control entirely via gaze for Cartesian robot control without the need for eye or face gestures.

Explore related subjects

cs.ROcs.SYeess.SY

Keep this discovery

BibTeXRIS

Emanuel Nunez Sardinha, Marcela Munera, Nancy Zook, David Western, Virginia Ruiz Garate. 2024-01-08. Diegetic Graphical User Interfaces and Intuitive Control of Assistive Robots via Eye-gaze. https://arxiv.org/abs/2401.03944

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

IMLE-VLA: Fast Single-Step Action Generation for Vision-Language-Action Policies

Vision-language-action (VLA) policies leverage pretrained vision-language backbones to achieve strong cross-task generalization. A leading design couples this backbone with a dedicated continuous action head trained via diffusion or flow matching. However, such heads rely on iterative multi-step sampling, for example 10 Euler steps in $\pi_{0.5}$. This creates an inference bottleneck that produces stop-and-go movement in the robot and slower task completion. We introduce IMLE-VLA, which replaces the iterative action head with a single-step conditional generator trained via conditional Implicit Maximum Likelihood Estimation (cIMLE). The cIMLE objective promotes multimodal action coverage, avoiding the mode collapse of naive regression heads while eliminating multi-step sampling entirely. When IMLE-VLA is applied to $\pi_{0.5}$, it increases inference frequency 3.67x (55 Hz vs. 15 Hz), enabling up to 11x higher action throughput. On the 40-task LIBERO benchmark, IMLE-VLA achieves the highest average success rate (98.0%) among all baselines while leading in inference frequency. Under the test-time perturbations of LIBERO-plus, IMLE-VLA retains $\pi_{0.5}$'s robustness while other baselines degrade sharply, confirming that the cIMLE head preserves generalization. Real-world experiments on a Franka Emika Panda across four tasks demonstrate smoother motion (2.2x to 3.0x lower jerk) and faster task completion, with IMLE-VLA outperforming $\pi_{0.5}$ on every task and reducing average VLA inference time per episode by 3.9x to 6.6x. Videos and code are available at https://kianhk6.github.io/IMLE-VLA/

cs.RO↗
arXiv2026

ObstaDiff: Generalizable Diffusion Policy Learning via Obstacle-aware Representations

Imitation learning has achieved impressive results in robotic manipulation, yet most existing approaches assume clean backgrounds and lack explicit mechanisms for obstacle-aware motion generation. Extending such policies to cluttered, real-world scenes with unstructured obstacles remains a key generalization challenge. We present ObstaDiff, a decomposed diffusion-policy framework with a lightweight obstacle-aware visual encoder. ObstaDiff extracts a structured target-obstacle-background representation, enabling the downstream alignment policy to generate end-effector trajectories toward a target-centered bottleneck pose while reasoning about surrounding obstacles. We evaluate ObstaDiff on 61 real-robot greenhouse trials per method (366 executions in total). ObstaDiff achieves 75.41% average task success and 8.20% average obstacle collision rate, outperforming representative imitation-learning baselines and improving generalization in cluttered agricultural scenes.

cs.RO↗
arXiv2026

Testing Between the Test Cases: Proving End-to-End Steering in Conditions You Never Drove

AI-based automated vehicle testing is challenging because a model that passes every test condition can still fail in the real world. Formal verification offers a way to directly address this gap. On a simulated highway and an arterial road we trained two small end-to-end steering networks each in CARLA, one on clear conditions alone and one on clear, fog, night and low sun. All four models were driven against a 2.19 ft lane-departure budget. Without driving again, we used bound propagation, a formal method that reads the trained weights, to compute how far steering can drift at every disturbance strength between two captured images. One calculation covers more than a campaign could drive: on the arterial it spans 133 poses, where ten intensities each would be 10^133 combinations, in minutes on one GPU. Not only did formal verification find conditions that broke the clear-trained policy without simulation testing, it provided some preliminary evidence for potential failures between the test cases. Our overall conclusion is that formal verification is a viable complement to simulation, and could be adopted as a part of verification and validation for automated driving.

cs.RO↗