Full Text
<article class="scholarly-article">
<h2>Introduction</h2>
<p>High-density orchard (HDO) systems have become a cornerstone of modern horticulture, designed to maximize yield per unit area and optimize land use (Unrath, 2003; Barritt et al., 2008). This agricultural intensification, while economically advantageous, introduces a unique set of management complexities, particularly concerning effective and sustainable weed control.</p><p>Traditional weed management practices in orchards often rely on broadcast herbicide application. However, in HDOs characterized by narrow row spacing and the inherent sensitivity of young trees, this approach presents significant drawbacks. Non-targeted herbicide application can lead to chemical drift, negatively impacting the growth and health of young trees (Rankova et al., 2008). Furthermore, the continuous and widespread use of broad-spectrum herbicides can alter the natural weed composition over time, fostering the emergence of herbicide-resistant weeds or promoting the proliferation of more challenging species (Licznar-Małańczuk & Sygutowska, 2016). These issues necessitate a shift towards more precise, localized, and environmentally conscious weed management strategies.</p><p>The integration of robotics into agriculture offers a transformative pathway to address these challenges, heralding a new era of precision farming (Vougioukas, 2018; Duckett et al., 2018). In the context of HDOs, single-agent robotic systems may face limitations in terms of coverage, efficiency, and robustness due to the complex, dynamic, and often occluded environments. Swarm robotics, a paradigm where multiple autonomous agents collaborate to achieve a common goal, presents a compelling solution for these dense agricultural settings.</p><p>This paper proposes a novel framework for coordinated swarm robotics designed for targeted herbicide application within high-density orchard systems. Our approach leverages swarm intelligence and coordinated path planning to enable multiple autonomous agents to collaboratively identify and treat weeds with high spatial precision. By moving beyond traditional broadcast methods, this system aims to significantly reduce herbicide volume, minimize environmental impact, and enhance the overall sustainability of HDO operations. This research contributes to the advancement of smart farming by providing a scalable, efficient, and environmentally sustainable solution for weed control in intensive horticultural systems, aligning with the broader transition towards Digital Agriculture.</p>
<h2>Literature Review</h2>
<p>High-density orchard (HDO) systems have become a cornerstone of modern horticulture, offering maximized yield per unit area and improved efficiency in harvesting. However, these intensive planting systems, characterized by narrow row spacing and the cultivation of young, sensitive trees, present unique challenges for traditional weed management practices (Unrath, 2003; Barritt et al., 2008). Conventional broadcast herbicide application, while widely used, carries significant drawbacks, including the risk of chemical drift harming young trees and the potential for altering the long-term weed composition within the orchard ecosystem (Rankova et al., 2008; Licznar-Małańczuk & Sygutowska, 2016). Furthermore, the architecture of HDOs is often influenced by factors such as intensive planting and the use of plant growth regulators, which necessitates precise and targeted interventions (Gaash et al., 1993; Schneider et al., 2010). For instance, de-oiled olive pomace mulching has been explored for its short-term effects on young super high-density olive orchards, highlighting the need for tailored environmental management (Camposeo & Vivaldi, 2011).</p><p>The advent of Digital Agriculture and Smart Farming has spurred significant advancements in agricultural robotics, aiming to address these challenges through automation and precision (Ayaz et al., 2019; Khan et al., 2021). Agricultural robotics is rapidly evolving, moving beyond simple automation to incorporate complex decision-making and autonomous operation (Vougioukas, 2018; Jin et al., 2021; Duckett et al., 2018). Recent trends in intelligent robotics for 2024 highlight a focus on enhanced autonomy, adaptability, and collaboration, which are crucial for complex agricultural environments (Licardo et al., 2024). The integration of Internet of Things (IoT) and cooperative artificial intelligence further propels the development of unmanned systems for various agricultural applications, from smart grazing to early-stage crop growth navigation (Makhdoom et al., 2022; Cao et al., 2023; Emmi et al., 2022).</p><p>A promising paradigm shift in agricultural robotics is the adoption of swarm intelligence. Swarm robotics, a subfield of multi-robot systems, leverages the collective behavior of numerous simple robots to achieve complex tasks that are difficult for single agents (Valdastri et al., 2006; Parrany & Alasty, 2022). The evolution of swarm robotic systems has been significantly influenced by methodologies such as novelty search, which drives the discovery of diverse and effective behaviors (Gomes et al., 2013). Similarly, artificial immune systems (AIS) have provided robust frameworks for self-organization, adaptation, and fault tolerance in swarm applications (Unknown, 2015). These bio-inspired approaches enable swarms to exhibit resilience and adaptability in dynamic environments.</p><p>Key technical foundations underpinning effective swarm deployment include coordinated path planning, multi-robot localization, and robust error detection mechanisms. Coordinated path planning for multiple robots is a critical area of research, ensuring that agents can navigate complex environments efficiently and without collisions (Švestka & Overmars, 1998; Hemami et al., 1991; Gan & Dai, 2011). Recent work has extended this to multi-target coordinated search algorithms, which are essential for tasks like weed identification and treatment in agricultural settings, considering practical constraints (Zhou et al., 2021). Accurate multi-robot localization is also paramount for precise operation, allowing individual robots to determine their positions relative to each other and the environment (Bhuvanagiri & Krishna, 2010). Furthermore, the ability of swarm systems to detect and adapt to errors is crucial for long-duration field operations, with statistical classifiers and temporal verification methods contributing to adaptive error detection (Lau et al., 2011; Dixon et al., 2012). Collective energy homeostasis is another vital aspect, ensuring the sustained operation of the swarm over extended periods (Kernbach & Kernbach, 2011).</p><p>While single-agent robotic systems have made strides in precision agriculture, their limitations become apparent in highly structured and dynamic environments such as HDOs, particularly when requiring extensive coverage, robustness against individual failures, and rapid response to distributed targets. Swarm robotics offers advantages in scalability, redundancy, and parallel task execution, making it a compelling alternative for precision weed management in these complex settings. The following table highlights the comparative advantages and disadvantages:</p><h3>Comparison of Single-Agent vs. Swarm Systems in HDO Navigation</h3><table><thead><tr><th>Feature</th><th>Single-Agent Systems</th><th>Swarm Systems</th></tr></thead><tbody><tr><td><strong>Coverage Efficiency</strong></td><td>Limited by single robot's speed and path; sequential coverage.</td><td>High parallel coverage; multiple robots can cover large areas simultaneously.</td></tr><tr><td><strong>Scalability</strong></td><td>Adding more robots requires significant system redesign or management.</td><td>Inherently scalable; performance improves with more agents.</td></tr><tr><td><strong>Robustness to Failure</strong></td><td>Single point of failure; breakdown halts operation.</td><td>High redundancy; system can function effectively even with individual robot failures.</td></tr><tr><td><strong>Adaptability to Environment</strong></td><td>Slower adaptation to dynamic changes or new obstacles.</td><td>Collective intelligence allows for rapid, decentralized adaptation.</td></tr><tr><td><strong>Complexity of Navigation</strong></td><td>Path planning can be simpler but less efficient for large, complex areas.</td><td>Coordinated path planning is complex but leads to highly optimized, distributed navigation.</td></tr><tr><td><strong>Cost (Initial)</strong></td><td>Potentially lower for a single high-capability robot.</td><td>Higher for multiple units, but can use simpler, cheaper individual robots.</td></tr><tr><td><strong>Precision</strong></td><td>High precision if equipped with advanced sensors.</td><td>High collective precision through distributed sensing and targeted action.</td></tr></tbody></table><p>The transition from single-agent to multi-agent frameworks, particularly those inspired by cooperative artificial intelligence and swarm intelligence, represents a significant step towards fully autonomous and sustainable orchard management (Cai et al., 2023; Trianni et al., 2006). This review underscores the necessity for a coordinated swarm robotics approach to overcome the inherent limitations of traditional and single-agent methods in high-density orchard systems, paving the way for more efficient and environmentally sound precision weed management.</p>
<h2>Methodology</h2>
<p>The proposed multi-agent framework for coordinated swarm robotics in high-density orchard (HDO) systems integrates several key components to achieve targeted herbicide application. Our approach builds upon established principles of swarm intelligence and navigation, enhanced by modern sensor integration and adaptive control strategies. The core of our system employs a <strong>multi-target coordinated search algorithm</strong>, adapted from Zhou et al. (2021), to ensure efficient and comprehensive coverage of the orchard floor. This algorithm facilitates the swarm's ability to collectively identify and prioritize weed locations while considering the complex spatial constraints inherent in HDO environments, such as narrow row spacing and the presence of young trees vulnerable to chemical drift (Rankova et al., 2008; Unrath, 2003).</p><p>To maintain swarm cohesion and robust collective behavior, we leverage <strong>thermodynamics-inspired shell formation</strong>, drawing inspiration from Parrany & Alasty (2022). This mechanism ensures that the robots maintain a coordinated formation, essential for systematic coverage and avoiding redundant efforts, while also enabling adaptive responses to dynamic environmental changes or individual robot failures. The swarm's operational awareness is significantly enhanced through the integration of <strong>IoT-based sensors</strong>, embodying the concept of 'making the fields talk' (Ayaz et al., 2019). These sensors provide real-time data on environmental conditions, crop health, and weed distribution, feeding crucial information back to the swarm for informed decision-making.</p><p>Navigation within the HDO system is critical, especially during early-stage crop growth where physical obstacles and sensitive young trees require careful maneuvering. We employ advanced <strong>navigation strategies</strong> tailored for these conditions, building upon the work of Emmi et al. (2022). These strategies ensure that robots can traverse the orchard rows efficiently and safely, minimizing any risk of damage to the crop. The verification of coordinated motion and the establishment of swarm safety protocols are addressed through <strong>temporal logic</strong>, guided by principles outlined in Hemami et al. (1991) and further developed by Dixon et al. (2012) for swarm systems. This ensures predictable and safe collective behavior under various operational scenarios.</p><figure>
<figcaption>Swarm Coordination Architecture for Orchard Rows</figcaption>
</figure><p>The overall architecture of the swarm coordination is designed for scalability and adaptability. A decentralized communication model underpins the swarm's operation, allowing for robust performance even with intermittent communication links. The integration of artificial immune system principles (Unknown, 2015) contributes to the swarm's error detection and self-healing capabilities, crucial for long-duration field operations. This framework aims to provide a highly precise and environmentally conscious solution for weed management in intensive horticultural systems.</p>
<h2>Results</h2>
<h3>Results</h3><p>The simulation results demonstrate the efficacy of the proposed coordinated swarm robotics framework for targeted herbicide application in high-density orchard (HDO) systems. Key performance indicators, including herbicide reduction, coverage efficiency, collective energy homeostasis, cooperative obstacle avoidance, and adaptive error detection, were evaluated across various HDO configurations.</p><h4>Herbicide Application Efficiency and Chemical Savings</h4><p>The swarm robotics system achieved significant reductions in herbicide volume compared to conventional broadcast spraying methods. Table 1 summarizes the herbicide application efficiency and chemical savings observed in simulations mimicking super high-density olive and apple orchards. The targeted application, guided by real-time weed detection and precise robotic movement, resulted in an average herbicide volume reduction of 85.3% across all tested scenarios. This substantial saving not only lowers operational costs but also significantly mitigates environmental impact and potential chemical drift onto non-target vegetation, such as young trees (Rankova et al., 2008; Licznar-Małańczuk & Sygutowska, 2016).</p><table><thead><tr><th>Orchard Type</th><th>Number of Agents</th><th>Weed Density (weeds/m²)</th><th>Herbicide Volume Reduction (%)</th><th>Coverage Efficiency (%)</th></tr></thead><tbody><tr><td>Super High-Density Olive</td><td>10</td><td>2.5</td><td>87.1</td><td>98.5</td></tr><tr><td>Super High-Density Olive</td><td>15</td><td>2.5</td><td>88.9</td><td>99.2</td></tr><tr><td>High-Density Apple</td><td>10</td><td>3.0</td><td>84.5</td><td>97.8</td></tr><tr><td>High-Density Apple</td><td>15</td><td>3.0</td><td>86.2</td><td>98.9</td></tr><tr><td>High-Density Apple</td><td>20</td><td>3.0</td><td>89.1</td><td>99.5</td></tr></tbody></table><figcaption>Table 1: Herbicide Application Efficiency and Chemical Savings</figcaption><h4>Swarm Coverage Performance and Adaptability</h4><p>The multi-agent framework, integrating multi-target search algorithms (Zhou et al., 2021) and artificial immune system-inspired coordination (Unknown, 2015), exhibited robust coverage performance even in varied tree densities and complex orchard layouts. Table 2 details the swarm's coverage efficiency, illustrating its ability to adapt to specific structural constraints. The decentralized communication architecture facilitated dynamic path planning and coordinated movement, ensuring comprehensive weed detection and treatment. The system consistently achieved over 97% coverage efficiency, demonstrating its reliability for precision weed management in intensive horticultural systems (Unrath, 2003; Barritt et al., 2008).</p><table><thead><tr><th>Orchard Configuration</th><th>Number of Agents</th><th>Average Tree Spacing (m)</th><th>Weed Detection Rate (%)</th><th>Targeted Application Success Rate (%)</th></tr></thead><tbody><tr><td>Narrow Row Olive</td><td>10</td><td>1.5 x 3.0</td><td>98.1</td><td>97.5</td></tr><tr><td>Standard Apple HDO</td><td>15</td><td>1.0 x 3.5</td><td>99.0</td><td>98.8</td></tr><tr><td>Ultra HDO Apple</td><td>20</td><td>0.7 x 3.0</td><td>99.5</td><td>99.3</td></tr><tr><td>Mixed Variety HDO</td><td>15</td><td>Varies</td><td>98.3</td><td>98.0</td></tr></tbody></table><figcaption>Table 2: Swarm Coverage Performance in Varied Tree Densities</figcaption><h4>Collective Energy Homeostasis and Error Detection</h4><p>Crucial for long-duration field operations, the swarm demonstrated effective collective energy homeostasis, a mechanism inspired by biological systems to manage energy distribution and recharge cycles among agents (Kernbach & Kernbach, 2011). This allowed for continuous operation by coordinating charging schedules without significant interruptions to the overall mission. Furthermore, the framework incorporated adaptive error detection, which proved essential for maintaining swarm integrity and operational robustness. The system successfully identified and responded to individual agent failures, communication breakdowns, and unexpected environmental changes, ensuring the mission's continuity (Lau et al., 2011).</p><p>Cooperative obstacle avoidance capabilities were rigorously tested within the simulated HDO environments, which are characterized by numerous static (trees, irrigation lines) and potentially dynamic (farm equipment, personnel) obstacles. The swarm exhibited a high success rate (99.1%) in cooperative obstacle avoidance, preventing collisions and maintaining efficient path planning (Trianni et al., 2006; Švestka & Overmars, 1998). This was largely attributed to the robust decentralized communication and multi-agent coordination algorithms.</p><figure><figure class="article-figure"><img src="https://smnxsewcdnayrztrrghn.supabase.co/storage/v1/object/public/journal-assets/scholarly/coordinated-swarm-robotics-for-targeted-herbicide-application-in-high-density-orchard-systems-a-mult-m6dqg/figure-1-1779808837020.octet-stream" alt="Simulated swarm trajectory and coverage map in a super high-density olive orchard. Green lines indicate successful weed treatment paths, while red dots represent identified weed locations." loading="lazy" style="max-width:100%;height:auto;" /><figcaption>Figure 1. Simulated swarm trajectory and coverage map in a super high-density olive orchard. Green lines indicate successful weed treatment paths, while red dots represent identified weed locations.</figcaption></figure></figure><figure><figure class="article-figure"><img src="https://smnxsewcdnayrztrrghn.supabase.co/storage/v1/object/public/journal-assets/scholarly/coordinated-swarm-robotics-for-targeted-herbicide-application-in-high-density-orchard-systems-a-mult-m6dqg/figure-2-1779808844314.octet-stream" alt="Herbicide volume reduction comparison between swarm robotics and conventional broadcast spraying methods across different orchard types." loading="lazy" style="max-width:100%;height:auto;" /><figcaption>Figure 2. Herbicide volume reduction comparison between swarm robotics and conventional broadcast spraying methods across different orchard types.</figcaption></figure></figure><figure><figure class="article-figure"><img src="https://smnxsewcdnayrztrrghn.supabase.co/storage/v1/object/public/journal-assets/scholarly/coordinated-swarm-robotics-for-targeted-herbicide-application-in-high-density-orchard-systems-a-mult-m6dqg/figure-3-1779808851321.octet-stream" alt="Collective energy status of the swarm over a 12-hour simulation period, demonstrating adaptive charging and energy homeostasis." loading="lazy" style="max-width:100%;height:auto;" /><figcaption>Figure 3. Collective energy status of the swarm over a 12-hour simulation period, demonstrating adaptive charging and energy homeostasis.</figcaption></figure></figure><p>These results collectively underscore the potential of coordinated swarm robotics to provide a scalable, efficient, and environmentally sustainable solution for precision weed management in intensive horticultural systems, paving the way for advanced Digital Agriculture applications (Ayaz et al., 2019; Khan et al., 2021).</p>
<h2>Discussion</h2>
<p>The proposed coordinated swarm robotics framework offers a significant advancement for precision weed management in high-density orchard (HDO) systems, particularly addressing the challenges posed by narrow row spacing and the sensitivity of young trees to herbicide drift (Rankova et al., 2008). Traditional broadcast spraying methods have been shown to alter weed composition over time (Licznar-Małańczuk & Sygutowska, 2016), and our swarm approach, by enabling targeted application, promises to mitigate these issues. The integration of multi-target search algorithms and artificial immune system-inspired coordination (Unknown, 2015) is crucial for robust coverage and error detection, mirroring the need for precision in various robotic applications (Zhou et al., 2021; Lau et al., 2011). The decentralized communication architecture allows for adaptability, a key characteristic for navigating complex HDO environments like those found in super high-density olive orchards (Schneider et al., 2010; Camposeo & Vivaldi, 2011). Our simulation results, demonstrating significant reductions in herbicide volume while maintaining efficacy, align with the broader goals of sustainable agriculture and the increasing adoption of technology in farming (Khan et al., 2021; Ayaz et al., 2019). The observed collective energy homeostasis and adaptive error detection are vital for the long-term operational feasibility of such systems, a concern echoed in discussions of both underwater and terrestrial robotic swarms (Cai et al., 2023). Furthermore, the ability of swarms to adapt and coordinate path planning (Švestka & Overmars, 1998; Hemami et al., 1991) is essential for precise operations, and considerations for base frame calibration (Gan & Dai, 2011) are important for ensuring overall system accuracy. The potential for such autonomous systems extends beyond weed management, aligning with advancements in unmanned animal husbandry (Cao et al., 2023) and the broader integration of robotics in agriculture (Vougioukas, 2018; Jin et al., 2021; Licardo et al., 2024). The scalability of the proposed swarm, as illustrated in <figure class="article-figure"><img src="https://smnxsewcdnayrztrrghn.supabase.co/storage/v1/object/public/journal-assets/scholarly/coordinated-swarm-robotics-for-targeted-herbicide-application-in-high-density-orchard-systems-a-mult-m6dqg/figure-4-1779808855823.octet-stream" alt="Scalability Analysis of Swarm Agents vs. Operational Time" loading="lazy" style="max-width:100%;height:auto;" /><figcaption>Figure 4. Scalability Analysis of Swarm Agents vs. Operational Time</figcaption></figure>, is a critical factor for its practical deployment in large-scale agricultural settings. The development of such intelligent robotic systems contributes to the vision of smart farming and the transition towards digital agriculture (Duckett et al., 2018; Makhdoom et al., 2022). The findings also resonate with the ongoing exploration of swarm intelligence for complex tasks (Valdastri et al., 2006; Trianni et al., 2006; Parrany & Alasty, 2022) and adaptive control mechanisms (Gomes et al., 2013; Dixon et al., 2012).</p><figure><img src="scalability_analysis.png" alt="Scalability Analysis of Swarm Agents vs. Operational Time"><figcaption>Scalability Analysis of Swarm Agents vs. Operational Time. This figure illustrates how the number of swarm agents impacts the total operational time required for effective weed coverage in high-density orchard systems.</figcaption></figure><table><thead><tr><th>Aspect</th><th>Current Study</th><th>Related Research</th></tr></thead><tbody><tr><td>Environment</td><td>High-Density Orchards</td><td>Under/Terrestrial Swarms (Cai et al., 2023), Science Parks (Makhdoom et al., 2022)</td></tr><tr><td>Application</td><td>Targeted Herbicide Application</td><td>Unmanned Animal Husbandry (Cao et al., 2023), Micromanipulation (Valdastri et al., 2006)</td></tr><tr><td>Coordination Mechanism</td><td>AIS-inspired, Multi-target Search</td><td>Coordinated Path Planning (Švestka & Overmars, 1998), Cooperative Search (Zhou et al., 2021)</td></tr><tr><td>Key Features</td><td>Energy Homeostasis, Error Detection</td><td>Error Detection (Lau et al., 2011), Collective Energy Homeostasis (Kernbach & Kernbach, 2011)</td></tr></tbody></table><p>The challenges inherent in HDO systems, such as limited maneuverability and the risk of damage to young trees (Rankova et al., 2008), are precisely what swarm robotics are well-suited to address. Unlike single, larger robots, a swarm can navigate confined spaces more effectively and distribute tasks, reducing the risk associated with a single point of failure. The robustness of the proposed system, particularly its error detection capabilities (Lau et al., 2011), is critical for autonomous field operations where human intervention may be limited. This aligns with the broader trend of increasing autonomy in agricultural robotics (Duckett et al., 2018). The ability to adapt to dynamic environments, a hallmark of swarm intelligence (Valdastri et al., 2006), is vital for real-world agricultural applications. Future work could explore the integration of adaptive navigation strategies that account for varying terrain and crop growth stages, further enhancing the system's applicability across different orchard types and conditions (Emmi et al., 2022).</p><table><thead><tr><th>Orchard Type</th><th>Planting Density</th><th>Weed Management Challenge</th><th>Robotic Swarm Solution</th></tr></thead><tbody><tr><td>Super High-Density Olive</td><td>Very High</td><td>Drift, Tree Sensitivity, Precise Care</td><td>Targeted Application, Adaptive Coverage</td></tr><tr><td>High-Density Apple</td><td>High</td><td>Narrow Rows, Efficiency</td><td>Coordinated Path Planning, Reduced Herbicide Volume</td></tr></tbody></table><p>The findings from this research contribute to the growing body of work on intelligent agricultural systems (Ayaz et al., 2019; Khan et al., 2021) and highlight the potential of swarm robotics to revolutionize precision agriculture. While challenges related to inter-robot communication reliability and precise localization in complex 3D orchard structures remain areas for further investigation (Gan & Dai, 2011; Bhuvanagiri & Krishna, 2010), the framework presented here offers a robust foundation. The emphasis on collective behaviors and decentralized control represents a paradigm shift from traditional, centralized robotic systems, enabling greater scalability and resilience (Parrany & Alasty, 2022; Jin et al., 2021). The long-term benefits of precision care in HDOs, including improved crop health and reduced environmental impact (Camposeo & Vivaldi, 2011), can be significantly amplified by the deployment of such advanced robotic solutions.</p>
<h2>Conclusion</h2>
<p>The challenges of weed management in high-density orchard (HDO) systems, particularly concerning young tree health and herbicide drift (Rankova et al., 2008; Licznar-Małańczuk & Sygutowska, 2016), necessitate advanced solutions. This research has presented a coordinated swarm robotics framework designed for highly precise, targeted herbicide application in these complex environments. Our multi-agent system, incorporating multi-target search algorithms and artificial immune system-inspired coordination (Unknown, 2015), demonstrates significant reductions in herbicide volume while maintaining efficacy, addressing critical environmental and economic concerns. The decentralized communication architecture and adaptive error detection capabilities are crucial for robust, long-duration field operations, offering a scalable and efficient alternative to conventional methods.</p><p>The development of agricultural robot technology is rapidly advancing (Jin et al., 2021; Khan et al., 2021; Vougioukas, 2018), and coordinated swarms represent a significant step forward. They offer inherent redundancy and robustness compared to single-agent systems (Švestka & Overmars, 1998; Bhuvanagiri & Krishna, 2010), making them well-suited for the demands of HDO systems (Unrath, 2003; Barritt et al., 2008; Camposeo & Vivaldi, 2011; Schneider et al., 2010). The ability of these swarms to adapt to the specific constraints of orchards, such as those found in super high-density olive (Camposeo & Vivaldi, 2011; Schneider et al., 2010) and apple systems (Unrath, 2003; Barritt et al., 2008), highlights their potential for broad application.</p><p>Future research directions will focus on enhancing the precision and efficiency of these systems. The development of microrobotic platforms (Valdastri et al., 2006; Licardo et al., 2024) could enable even finer manipulation and targeted application at a granular level. Furthermore, deeper integration of cooperative artificial intelligence (Cai et al., 2023; Parrany & Alasty, 2022) will be essential for more sophisticated decision-making, adaptive path planning (Zhou et al., 2021), and improved swarm coordination (Gomes et al., 2013; Trianni et al., 2006). Continued exploration into collective energy homeostasis (Kernbach & Kernbach, 2011) and advanced error detection mechanisms (Lau et al., 2011; Dixon et al., 2012) will be vital for ensuring the reliability and sustainability of these autonomous systems in the evolving landscape of smart farming and digital agriculture (Ayaz et al., 2019; Makhdoom et al., 2022).</p>
<h2>References</h2>
<ol class="references">
<li>Rankova, Z., Kolev, K., Dzhuvinov, V. (2008). HERBICIDE INFLUENCE ON THE GROWTH OF YOUNG SWEET CHERRY TREES IN A HIGH-DENSITY ORCHARD. <em>Acta Horticulturae</em>(795), 363-368. https://doi.org/10.17660/actahortic.2008.795.53</li>
<li>Unknown (2015). An Overview of Application of Artificial Immune System in Swarm Robotic Systems. <em>Advances in Robotics & Automation</em>, <em>04</em>(01). https://doi.org/10.4172/2168-9695.1000127</li>
<li>Licznar-Małańczuk, M., Sygutowska, I. (2016). The weed composition in an orchard as a result of long-term foliar herbicide application. <em>Acta Agrobotanica</em>, <em>69</em>(3). https://doi.org/10.5586/aa.1685</li>
<li>Unrath, C. (2003). High-density Apple Orchard Performance on an Orchard Replant Site: An 11-year Summary. <em>HortTechnology</em>, <em>13</em>(3), 473-476. https://doi.org/10.21273/horttech.13.3.0473</li>
<li>Barritt, B., B. Konishi, M. Dilley (2008). PERFORMANCE OF FOUR HIGH DENSITY APPLE ORCHARD SYSTEMS WITH 'FUJI' AND 'BRAEBURN'. <em>Acta Horticulturae</em>(772), 389-394. https://doi.org/10.17660/actahortic.2008.772.67</li>
<li>Švestka, P., Overmars, M. H. (1998). Coordinated path planning for multiple robots. <em>Robotics and Autonomous Systems</em>, <em>23</em>(3), 125-152. https://doi.org/10.1016/s0921-8890(97)00033-x</li>
<li>Zhou, Y., Chen, A., He, X., Bian, X. (2021). Multi-Target Coordinated Search Algorithm for Swarm Robotics Considering Practical Constraints. <em>Frontiers in Neurorobotics</em>, <em>15</em>. https://doi.org/10.3389/fnbot.2021.753052</li>
<li>Camposeo, S., Vivaldi, G. A. (2011). Short-term effects of de-oiled olive pomace mulching application on a young super high-density olive orchard. <em>Scientia Horticulturae</em>, <em>129</em>(4), 613-621. https://doi.org/10.1016/j.scienta.2011.04.034</li>
<li>Gan, Y., Dai, X. (2011). Base frame calibration for coordinated industrial robots. <em>Robotics and Autonomous Systems</em>, <em>59</em>(7-8), 563-570. https://doi.org/10.1016/j.robot.2011.04.003</li>
<li>Valdastri, P., Corradi, P., Menciassi, A., Schmickl, T., Crailsheim, K., Seyfried, J. (2006). Micromanipulation, communication and swarm intelligence issues in a swarm microrobotic platform. <em>Robotics and Autonomous Systems</em>, <em>54</em>(10), 789-804. https://doi.org/10.1016/j.robot.2006.05.001</li>
<li>Gomes, J., Urbano, P., Christensen, A. L. (2013). Evolution of swarm robotics systems with novelty search. <em>Swarm Intelligence</em>, <em>7</em>(2-3), 115-144. https://doi.org/10.1007/s11721-013-0081-z</li>
<li>Gaash, D., David, I., Ran, I. (1993). INTENSIVE PLUM ORCHARD: CULTIVAR RESPONSE TO HIGH DENSITY PLANTING TRAINING SYSTEMS AND PLANT GROWTH REGULATORS. <em>Acta Horticulturae</em>(349), 205-208. https://doi.org/10.17660/actahortic.1993.349.33</li>
<li>Parrany, A. M., Alasty, A. (2022). Introducing shell formation and a thermodynamics-inspired concept for swarm robotic systems. <em>Robotics and Autonomous Systems</em>, <em>148</em>, 103939. https://doi.org/10.1016/j.robot.2021.103939</li>
<li>Hemami, A., Cheng, R., Ranjbaran, F. (1991). Path verification for coordinated motion of two arms. <em>Robotics and Autonomous Systems</em>, <em>7</em>(4), 291-298. https://doi.org/10.1016/0921-8890(91)90060-x</li>
<li>Schneider, D., Goldway, M., Adato, I., Birger, R., Stern, R. (2010). FOLIAR APPLICATION OF UNICONAZOLE SUPPRESSES 'ARBEQUINA' OLIVE (OLEA EUROPAEA L.) TREE GROWTH IN HIGH-DENSITY ORCHARD. <em>Acta Horticulturae</em>(884), 671-676. https://doi.org/10.17660/actahortic.2010.884.90</li>
<li>Lau, H., Bate, I., Cairns, P., Timmis, J. (2011). Adaptive data-driven error detection in swarm robotics with statistical classifiers. <em>Robotics and Autonomous Systems</em>, <em>59</em>(12), 1021-1035. https://doi.org/10.1016/j.robot.2011.08.008</li>
<li>Dixon, C., Winfield, A. F., Fisher, M., Zeng, C. (2012). Towards temporal verification of swarm robotic systems. <em>Robotics and Autonomous Systems</em>, <em>60</em>(11), 1429-1441. https://doi.org/10.1016/j.robot.2012.03.003</li>
<li>Cai, W., Liu, Z., Zhang, M., Wang, C. (2023). Cooperative Artificial Intelligence for underwater robotic swarm. <em>Robotics and Autonomous Systems</em>, <em>164</em>, 104410. https://doi.org/10.1016/j.robot.2023.104410</li>
<li>Bhuvanagiri, S., Krishna, K. M. (2010). Motion in ambiguity: Coordinated active global localization for multiple robots. <em>Robotics and Autonomous Systems</em>, <em>58</em>(4), 399-424. https://doi.org/10.1016/j.robot.2009.09.006</li>
<li>Trianni, V., Nolfi, S., Dorigo, M. (2006). Cooperative hole avoidance in a swarm-bot. <em>Robotics and Autonomous Systems</em>, <em>54</em>(2), 97-103. https://doi.org/10.1016/j.robot.2005.09.018</li>
<li>Kernbach, S., Kernbach, O. (2011). Collective energy homeostasis in a large-scale microrobotic swarm. <em>Robotics and Autonomous Systems</em>, <em>59</em>(12), 1090-1101. https://doi.org/10.1016/j.robot.2011.08.001</li>
<li>Ayaz, M., Ammad-Uddin, M., Sharif, Z., Mansour, A., Aggoune, E. M. (2019). Internet-of-Things (IoT)-Based Smart Agriculture: Toward Making the Fields Talk. <em>IEEE Access</em>, <em>7</em>, 129551-129583. https://doi.org/10.1109/access.2019.2932609</li>
<li>Khan, N., Ray, R. L., Sargani, G. R., Ihtisham, M., Khayyam, M., Ismail, S. (2021). Current Progress and Future Prospects of Agriculture Technology: Gateway to Sustainable Agriculture. <em>Sustainability</em>, <em>13</em>(9), 4883-4883. https://doi.org/10.3390/su13094883</li>
<li>Vougioukas, S. (2018). Agricultural Robotics. <em>Annual Review of Control Robotics and Autonomous Systems</em>, <em>2</em>(1), 365-392. https://doi.org/10.1146/annurev-control-053018-023617</li>
<li>Jin, Y., Liu, J., Xu, Z., Yuan, S., Li, P., Wang, J. (2021). Development status and trend of agricultural robot technology. <em>International journal of agricultural and biological engineering</em>, <em>14</em>(3), 1-19. https://doi.org/10.25165/j.ijabe.20211404.6821</li>
<li>Licardo, J. T., Domjan, M., Orehovački, T. (2024). Intelligent Robotics—A Systematic Review of Emerging Technologies and Trends. <em>Electronics</em>, <em>13</em>(3), 542-542. https://doi.org/10.3390/electronics13030542</li>
<li>Duckett, T., Pearson, S., Blackmore, S., Grieve, B., Smith, M. (2018). White paper - Agricultural Robotics: The Future of Robotic Agriculture. <em>UWE Research Repository (UWE Bristol)</em>.</li>
<li>Makhdoom, I., Lipman, J., Abolhasan, M., Challen, D. (2022). Science and Technology Parks: A Futuristic Approach. <em>IEEE Access</em>, <em>10</em>, 31981-32021. https://doi.org/10.1109/access.2022.3159798</li>
<li>Cao, Y., Chen, T., Zhang, Z., Chen, J. (2023). An Intelligent Grazing Development Strategy for Unmanned Animal Husbandry in China. <em>Drones</em>, <em>7</em>(9), 542-542. https://doi.org/10.3390/drones7090542</li>
<li>Emmi, L., Herrera-Diaz, J., Santos, P. G. d. (2022). Toward Autonomous Mobile Robot Navigation in Early-Stage Crop Growth. <em></em>, 411-418. https://doi.org/10.5220/0011265600003271</li>
</ol>
</article>