Operating an agricultural drone is a complex task that demands constant attention and rapid decision-making. The interface serves as the primary medium for information exchange between the operator and the machine. The operator’s situational awareness (SA)—their perception, comprehension, and projection of the drone’s state and environment—directly impacts both operational efficiency and safety. Consequently, the design of the agricultural drone interface is a critical factor in system performance. With advancements in digital media and display technologies, the visual design of these digital interfaces has garnered significant attention for its ergonomic impact. However, the vast amount of data presented can overwhelm the ground operator, increasing cognitive load, delaying decisions, and potentially leading to accidents. This research, therefore, focuses on applying situational awareness theory to the optimization of an agricultural drone operation interface. The goal is to develop methods for interface interaction design that enhance the operator’s SA, thereby improving work performance and safety. The findings are intended to provide a reference for optimizing operator interfaces in agricultural drones and other engineering machinery.
The theoretical foundation of this work is rooted in the concept of Situational Awareness. In the context of human-machine systems, SA is defined as the perception of environmental elements within a volume of time and space, the comprehension of their meaning, and the projection of their status in the near future. This established three-level model provides a robust framework for analysis. A well-designed interface must support all three levels to enable effective operator performance. The relationship between interface design, SA, and operator action can be conceptually modeled, where poor design directly degrades SA, leading to suboptimal decisions. A significant challenge in complex systems like agricultural drone operation is information overload. The interface must present critical data without overwhelming the operator. This challenge can be represented by a simple relationship:
$$I_o = \frac{I_p}{C_p}$$
where $I_o$ is the perceived information overload, $I_p$ is the information presented by the interface, and $C_p$ is the operator’s cognitive processing capacity. Effective interface design aims to minimize $I_o$ by strategically managing $I_p$ to align with $C_p$.
Research on applying SA theory to human-machine interfaces is extensive. Studies in aviation, nuclear power plant control rooms, and automotive contexts have demonstrated that ecological interface design (EID) and systematic evaluation methods can significantly improve operator SA. For instance, research has combined behavioral experiments with physiological measurements like EEG to objectively assess the “perceivability, comprehensibility, and projectability” of digital interfaces. Other work has developed multi-agent models to analyze SA breakdowns in accident scenarios. These principles are highly applicable to the domain of agricultural drone operation, where the operator must maintain SA over flight parameters, system status, payload data, and a dynamic environment.
Current research on agricultural drone interfaces often focuses on communication systems, flight control algorithms, and automation features. While some studies have implemented ground control stations with real-time telemetry and mapping, less attention has been paid to the cognitive ergonomics and SA-oriented design of the visual interface itself. Evaluations of controller usability based on different mental models have shown that pilot-centric designs yield higher performance, underscoring the importance of user-centered design. The optimization strategy proposed here directly addresses this gap by analyzing existing interfaces through the lens of SA theory.
Our analysis of a representative agricultural drone operation interface revealed several key problems that impair the operator’s SA. These issues primarily concern the visual presentation and organization of critical information elements, creating inefficiencies and potential safety risks. The table below summarizes the identified problems and their impact on the three levels of SA.
| Interface Element | Identified Problem | Impact on Situational Awareness |
|---|---|---|
| Obstacle Proximity Indicator | Visual form is not salient; distance is not intuitively clear. The operator cannot quickly judge the distance to an obstacle. | Level 1 (Perception): Hinders immediate detection of critical threats. Level 2 (Comprehension): Slows understanding of collision risk. |
| Drone Battery Level | Icon provides only an approximate graphical representation; precise numerical reading is difficult to ascertain. | Level 2 (Comprehension): Impedes accurate assessment of remaining operational time. Level 3 (Projection): Compromises ability to predict mission feasibility. |
| Battery Icons (Drone vs. Controller) | Inconsistent visual design between the drone battery and the remote controller battery icons. | Level 1 (Perception): Increases cognitive load by forcing interpretation of two different schemas. |
| Remaining Chemical/Liquid | Icon is small and positioned non-prominently; current level is not immediately noticeable. | Level 1 (Perception): Critical task information is easy to miss. Level 3 (Projection): Risk of running out of payload mid-operation without warning. |
Based on these findings and grounded in SA theory, we propose three core design principles for optimizing the agricultural drone operation interface:
1. Clarity and Simplicity: The interface must prioritize legibility and intuitive understanding. Information should be presented in an easily digestible format to minimize the time and effort required for perception and comprehension. This reduces the likelihood of operator error, which is crucial for safe agricultural drone operation in unpredictable field conditions.
2. Consistency: A consistent interface allows the operator to build and rely on stable mental models. This applies to visual style (colors, fonts, layout structures), interaction patterns, and, critically, the semantic meaning of visual cues. For example, identical iconography and color codes should represent analogous states (e.g., battery levels) across different system components.
3. Learnability: Given the diverse backgrounds of agricultural drone operators, the interface should be easy to learn and remember. Design choices should leverage common conventions and affordances to make the system self-explanatory where possible, reducing training overhead and promoting correct usage.
Guided by these principles and the SA framework, we developed targeted optimization strategies for the problematic elements. The core idea is to enhance information salience and semantics through intelligent visual coding, primarily using color. We established a three-tier color semantics system: Red for dangerous/critical states requiring immediate attention, Yellow/Amber for cautionary/warning states, and Green for normal/safe states. This system directly supports SA by enabling rapid perception (Level 1) and immediate comprehension of state severity (Level 2).
The optimized design for each element is as follows:
1. Obstacle Proximity Visualization: The obstacle icon’s color now dynamically changes based on distance, accompanied by a precise numerical distance and time-to-contact estimate. The logic is formalized as:
$$
\text{Color}_{\text{obstacle}} =
\begin{cases}
\text{Red}, & \text{if } d \leq 8\text{m} \\
\text{Yellow}, & \text{if } 8\text{m} < d \leq 20\text{m} \\
\text{Green}, & \text{if } d > 20\text{m}
\end{cases}
$$
where $d$ is the distance to the obstacle. This allows the agricultural drone operator to assess collision risk at a glance.
2 & 3. Unified Battery Status Display: The drone and remote controller battery icons were redesigned to have a consistent visual form. Both now feature a clear numerical percentage readout. Furthermore, the drone battery icon implements color semantics:
$$
\text{Color}_{\text{battery}} =
\begin{cases}
\text{Red}, & \text{if } \text{charge} \leq 10\% \\
\text{Yellow}, & \text{if } 10\% < \text{charge} \leq 30\% \\
\text{Green}, & \text{if } \text{charge} > 30\%
\end{cases}
$$
This enables the operator of the agricultural drone to instantly understand not just the quantity, but the implications of the remaining charge.
4. Salient Chemical/Liquid Remaining Display: The remaining liquid icon was redesigned for visual prominence and aligned with the battery icon style for consistency. More importantly, it employs a similar color-coded logic based on operational thresholds. Assuming a standard tank capacity, the semantics could be:
$$
\text{Color}_{\text{liquid}} =
\begin{cases}
\text{Red}, & \text{if } V_{\text{remaining}} \leq 3.2\text{L} \\
\text{Yellow}, & \text{if } 3.2\text{L} < V_{\text{remaining}} \leq 9.6\text{L} \\
\text{Green}, & \text{if } V_{\text{remaining}} > 9.6\text{L}
\end{cases}
$$
where $V_{\text{remaining}}$ is the remaining liquid volume. This ensures the agricultural drone operator is proactively alerted to the need for refilling, preventing damage from dry running and mission interruption.
| Element | Optimization | SA Level Supported | Design Principle Applied |
|---|---|---|---|
| Obstacle Indicator | Dynamic color coding + precise distance/time data. | L1: Enhanced salience. L2: Clear risk comprehension. L3: Informed evasive action. |
Clarity, Consistency (in color semantics) |
| Battery Status | Unified iconography, numerical readout, color semantics. | L1: Quick perception of level/state. L2/L3: Accurate runtime understanding and projection. |
Consistency, Clarity, Learnability |
| Remaining Liquid | Prominent placement, consistent icon, color semantics. | L1: Harder to miss. L2/L3: Clear understanding of payload status and refill needs. |
Clarity, Consistency |
The implementation of these optimizations results in a cleaner, more informative interface. The key change is the transformation of static, quantitative displays into dynamic, semantically-rich visualizations. The operator is no longer required to mentally calculate or interpret raw numbers under time pressure. Instead, the interface itself performs an initial level of “comprehension” through its color-coding logic and presents the result in an immediately actionable format. This direct perception of state meaning is the essence of supporting high-level SA in agricultural drone operation. The before-and-after comparison, while not shown here, demonstrates a significant reduction in visual clutter and a marked improvement in the salience of critical flight and system parameters.
In conclusion, the interface for an agricultural drone is a vital cognitive tool, not merely a display panel. Its design must be treated as a systematic engineering effort centered on the operator’s cognitive needs and workflows. By applying the structured lens of situational awareness theory—focusing on perception, comprehension, and projection—we can identify specific deficiencies in existing interfaces. Guiding the redesign with principles of clarity, consistency, and learnability, and employing techniques like semantic color coding, leads to tangible optimizations. These changes empower the agricultural drone operator to build and maintain a more accurate and timely mental model of the operational environment. This enhanced SA is crucial for effective decision-making in challenging conditions like low visibility, adverse weather, or complex field geometries, ultimately contributing to safer and more efficient precision agriculture operations. Future work should involve rigorous user testing with target operators to quantitatively validate the improvements in SA, workload, and task performance, further refining the design principles for this critical domain.