Written by 10:54 am AI Trend

### High School Student Develops A.I.-Powered Device to Eliminate Aggressive Lanternflies

Using solar power, machine learning and her family’s patio umbrella, 18-year-old Selina Zhang creat…

A native of New Jersey, Selina Zhang is well acquainted with the spotted lanternfly, an invasive species that has wreaked havoc on the local agricultural industry in the Garden State for an extended period. Back when she was just a young girl of 18 years, Zhang first encountered this pest while visiting a market near her hometown of Annandale. She was captivated by a group of these vibrant, moth-like insects, with their striking appearance of bright red underwings contrasting against black polka-dotted forewings. However, despite their visually appealing exterior, the deceptive nature of the spotted lanternfly soon became apparent.

Belonging to the plant-hopper category, this insect leaps from one organism to another to feed on their sap, impacting over 70 different species in the process. As a result, the affected plants experience heightened stress levels, making them more vulnerable to diseases and attacks. Moreover, the spotted lanternfly’s consumption of large quantities of sugar leads to the excretion of honeydew, a sticky liquid waste that accumulates on trees. This accumulation impedes photosynthesis and fosters the growth of black sooty mold.

Originating from China, the spotted lanternfly likely made its way to the United States in 2012 concealed within a shipment of stones in the form of egg masses. The first significant sighting of a group of these plant-hoppers occurred in 2014 in Berks County, Pennsylvania, where they covered a wooden area, marking the beginning of a series of infestations across 17 states.

“As I matured, I felt compelled to take tangible action,” Zhang expressed. “I aimed to develop an innovative solution that leveraged my unique perspective and existing research to combat this pest in novel ways.”

Traditional methods of managing the spotted lanternfly typically involve the use of insecticides and sticky bands—both of which have adverse ecological repercussions. Insecticides, such as neonicotinoids, leave behind harmful residues that pose a threat to essential pollinators like honeybees, while sticky bands have the unintended consequence of ensnaring birds, small animals, and other insects.

To circumvent these negative impacts, Zhang drew inspiration from chess boards and the game “Dance Dance Revolution.” Through weeks of meticulous field observations, extensive algorithmic programming, and the creative repurposing of an umbrella from her family’s patio, she conceptualized ArTreeficial. This solar-powered, self-cleaning, artificial intelligence-driven “tree” lures the spotted lanternfly and eliminates the pest using an electronic mesh.

“In essence, ArTreeficial functions as a large-scale mousetrap designed to combat the invasive spotted lanternfly,” Zhang explained. “The bait used is primarily derived from the tree of heaven [Ailanthus altissima], a plant highly attractive to the spotted lanternfly.”

In addition to conducting a thorough review of existing literature, Zhang meticulously observed the behavior of the spotted lanternfly in its natural habitat over several weeks. She documented over 500 photographs of these insects, creating a personal database for reference.

“The primary goal of the field observation was to gain deeper insights into the behavior of the adversary. Without a comprehensive understanding of their habits, creating an effective trap would be challenging,” Zhang remarked.

Zhang’s vision involved creating a tree-like structure that mimicked the tree of heaven, a favored host plant of the spotted lanternfly, to serve as the primary lure for her trap. She repurposed the umbrella from her family’s patio to commence her work.

Initially, Zhang utilized ultrasound technology to emit an incense derived from the essence of the tree of heaven from ArTreeficial’s trunk to attract the pests. Subsequently, she designed an electric net for the synthetic tree, intended to shock spotted lanternflies upon contact. In the initial iteration, Zhang encountered two main challenges: the lack of a protective barrier around the electric net and the potential evasion of the single net by agile spotted lanternflies. Reflecting on her field observations, Zhang recalled the delicate, elongated strides taken by the insects, often extending and wiggling their spindly legs before landing.

Taking this behavior into account, Zhang upgraded to a double-layered net structure, incorporating a large diamond-shaped outer net with a tighter inner mesh onto ArTreeficial. This configuration allowed the outer layer to act as a safety shield, enabling the spotted lanternfly to pass through the small apertures of the upper mesh and make contact with the inner mesh.

“Field observations necessitate patience and perseverance, as they often involve prolonged periods of uneventful monitoring. However, they are crucial for yielding valuable insights,” Zhang emphasized.

The configuration of the double net had to conform to the trunk of ArTreeficial, presenting a challenge in maintaining the optimal distance between the two layers. To address this, Zhang used rubber tubing to interlace the nets at a fixed distance, ensuring effective functionality.

Subsequently, Zhang integrated artificial intelligence into her invention, leveraging the database of photographs to inform and program her AI model.

“I opted to rely on my personal collection of photographs rather than generic internet images, which often emphasize specific features of the spotted lanternfly, such as its vibrant red wings or clustered formations,” Zhang explained. “High-quality data is essential for training a robust AI model. My field observations and images provide a holistic understanding for the model to learn from.”

Similar to a chessboard, Zhang’s nets are segmented into multiple squares along the structure. When a spotted lanternfly lands on the inner net, the AI model is activated, delivering an electric charge to that specific section while the rest remain inactive. Consequently, the insect is electrocuted—a process akin to a lethal, reverse “Dance Dance Revolution” game. In this popular music video game, players step on a tile grid featuring directional arrows corresponding to dance steps displayed on the screen. The illuminated arrow tile signals the next move in the routine, prompting the player to execute the corresponding step. Zhang’s innovation follows a similar principle, albeit in a different sequence, with the electronic signal triggering after the step.

“It was truly exhilarating to witness the system in action,” Zhang shared. “The moment I activated the net and tested it with a screwdriver, witnessing the spark, was truly a eureka moment.”

Zhang has been selected as one of the 40 finalists in the Regeneron Science Talent Search, the nation’s esteemed science and math competition for high school seniors. The finalist projects span a wide spectrum, ranging from groundbreaking engineering accomplishments like low-cost, universal 3D-printed prosthetics for amputees to space science breakthroughs such as the analysis of microorganisms on comets and asteroids to speculate on the origins of life on Earth. Zhang’s project falls under the environmental science category of the competition, showcasing its interdisciplinary nature.

“The project integrates AI, chemistry, addresses climate change, and harnesses solar power. It embodies the interdisciplinary essence of science and engineering,” remarked Maya Ajmera, the president and CEO of the Society for Science, the organization behind the talent search. “This multifaceted approach is what distinguishes it in my eyes.”

Looking ahead, Zhang plans to explore how the essence extracted from deceased spotted lanternflies can enhance her lure and intends to conduct field tests to validate her invention’s efficacy.

At present, the cost of manufacturing ArTreeficial amounts to approximately $200. Zhang is actively engaging with local farmers and aims to collaborate with regional companies to streamline production and reduce costs over time. Her vision involves deploying multiple synthetic trees simultaneously and integrating them into a network, which she fondly refers to as “ArTreeForest.” Once the effectiveness of her system against spotted lanternflies is established, Zhang envisions extending its application to combat other invasive species.

“I aspire to develop ArTreeficial as a universal paradigm, not limited to addressing the spotted lanternfly issue but as a versatile tool applicable to pest control technologies,” Zhang articulated.

A distinguished violinist who has graced the stage at Carnegie Hall and an active participant in extracurricular activities like the Science Bowl and USA Biology Olympiad at North Hunterdon High School, Zhang exemplifies a fusion of talent and creativity. A fervent artist and writer, she even transformed the spotted lanternfly into a formidable character in a comic book, aiming to raise awareness about this local invasive species through a creative outlet.

“Wherever it spreads, disaster follows,” Zhang remarked. “Through my comics, I sought to foster a deeper understanding of this local pest and its behavior among the general public.”

As she approaches the culmination of her high school journey, Zhang sets her sights on college, particularly seeking a university renowned for its strong biology and computer science programs. This educational path will enable her to further integrate artificial intelligence into biological research endeavors.

“I aim to showcase how artificial resources can serve as indispensable tools in safeguarding our natural resources, particularly in agriculture,” Zhang asserted. “As stewards of the Earth, it is our duty to protect and nurture these resources responsibly, ensuring their preservation for present and future generations to cherish.”

Visited 4 times, 1 visit(s) today
Tags: Last modified: March 9, 2024
Close Search Window
Close