Enabling Humans to Plan Inspection Paths Using a Virtual Reality Interface

Abstract: In this work, we investigate whether humans can manually generate high-quality robot paths for optical inspections. Typically, automated algorithms are used to solve the inspection planning problem. The use of automated algorithms implies that specialized knowledge from users is needed to set up the algorithm. We aim to replace this need for specialized experience, by entrusting a non-expert human user with the planning task. We augment this user with intuitive visualizations and interactions in virtual reality. To investigate if humans can generate high-quality inspection paths, we perform a user study in which users from different experience categories, generate inspection paths with the proposed virtual reality interface. From our study, it can be concluded that users without experience can generate high-quality inspection paths: The median inspection quality of user generated paths ranged between 66-81\% of the quality of a state-of-the-art automated algorithm on various inspection planning scenarios. We noticed however, a sizable variation in the performance of users, which is a result of some typical user behaviors. These behaviors are discussed, and possible solutions are provided.

• The paper demonstrates that humans using a virtual reality interface can plan robotic inspection paths achieving 66-81% of automated algorithm quality while being significantly faster.
• A user study showed significant variability in human performance, with some users exceeding automated path quality but others hindered by sub-optimal strategies.
• The study suggests VR interfaces can democratize complex robotic inspection planning, reducing dependency on specialized expertise and enabling faster setup times.

Overview of the Paper on Virtual Reality-based Robotic Inspection Path Planning

The presented paper investigates the potential for human users, specifically those without specialized experience, to manually generate high-quality robotic paths for optical inspections using a virtual reality (VR) interface. The conventional approach to solving the inspection planning problem heavily relies on automated algorithms which, though effective, necessitate specialized knowledge in robotics and optical inspection for their setup. This research aims to democratize the path planning process by leveraging intuitive visualizations and interactions delivered through a VR environment, potentially enabling non-expert users to generate effective robotic inspection trajectories.

Objectives and Methodology

The primary objective was to evaluate whether non-expert users could produce inspection paths with quality comparable to that generated by state-of-the-art automated algorithms. Key tasks within the research included developing a VR interface that features interactive quality visualization and controlling robots using VR controllers. The VR system was then tested through a user study comprising participants with varying levels of expertise in robotics and inspection techniques.

To quantify the effectiveness of user-generated paths, a quality ratio was defined as the ratio of the inspection quality of paths created by users against those generated by an automated solution under the same budget constraints. The performance of users was assessed through six small-scale inspection scenarios and one complex, large-scale scenario.

Key Findings

1. User Performance: The median performance of human-generated inspection paths was reported to attain 66-81% of the quality of paths generated by a state-of-the-art automated algorithm. Certain users even exceeded automated path quality in complex scenarios, with quality ratios reaching 124%.
2. Time Efficiency: Users were significantly faster in generating inspection paths compared to automated systems. The maximum time taken by a user for generating a complex path was noted to be approximately 9 minutes, whereas the automated algorithm required several hours.
3. Variability and Challenges: A considerable variability in user performance was observed, highlighting the influence of user strategy and execution. Common pitfalls included sub-optimal positioning leading to increased path lengths and detours that didn't substantially contribute to measurement quality.

Implications and Future Directions

The study underscores the feasibility of utilizing a VR-based interface to reduce the dependency on specialized knowledge for inspection path planning. Practical implications include decreased setup time for inspection planning and lower operational demands on expertise, which can democratize complex robotic tasks in industry.

Theoretically, the research opens avenues for designing enhanced human-machine interfaces where intuitive user interaction can bridge gaps traditionally filled by algorithmic proficiency. This approach can potentially be extended to other domains requiring intricate path planning.

Future work could explore hybrid systems, integrating user intuition through VR interfaces with the precision and optimization capabilities of automated algorithms. Developing mechanisms to dynamically aid users in reducing path complexities, for instance, through AI-driven suggestions or post-processing enhancements, could further augment the quality and efficiency of user-driven paths.

Overall, the study presents a valuable exploration into human-centric approaches in robotics, demonstrating how VR technology can facilitate more accessible and efficient solutions in high-dimensional, complex tasks traditionally dominated by computational algorithms.