Lincoln Abraham Park Full Report With Emily Beach, Community Programs Manager at Chicago Parks Foundation

Report created by: EccoDrone 

Map and Key: 

Key: 

Blue markers: Plastic pieces 

Green markers: Strips of paper, receipts, wrappers, crumpled paper balls 

Red markers: Other bits of trash, tar, cans, bottles, cloth, bandaids, chips, tinfoil, fallen tree branches, etc.

Overview: 

Generally, most trash are concentrated in larger fields and areas. These areas are also closer to the main road, in this case, this is North Boulevard. North Boulevard also contains most of the park's openings and 2 main areas for sitting. Main clusters of papers and plastics are prominent around the edges of large park spaces, near roads and footpaths. Bottles and other larger trashes are concentrated around trees surrounding open park areas and La Salle Drive. This park does not contain large amounts of trash and the trash is fairly spread out and small.

Data, Charts, and Statistics 

Trash 

Types of Trash Collected: These are the types of trash that was detected through the video footage. Some of the trash may be labeled in a different category due to their shape and frequency. The most common types of trash were small scraps of paper. There were some fallen tree branches and abandoned scooters that may not be considered ‘trash’ but were still detected.

● Total objects detected: 126 objects detected

● Non trash detected: 3 

Trash Concentrations: 

● High Concentration: Around edges of large open spaces and near major road 

●High - Area 2, 3, 4, 5, 12 19 

● Medium Concentration: Smaller open 

spaces and around smaller roads 

●Medium - Area 1, 6, 18, 23 

● Low Concentration: Centers of large 

open spaces and small open spaces 

●Low - Area 7, 8, 9, 10, 11, 13, 14, 15, 16, 17, 20, 21, 22 

Papers: 64 trash pieces 

Plastics: 35 trash pieces 

Other: 24 trash pieces 

Bottles/cans: 13 

Trees: 2 

Scooter: 1 

Cloth: 2 

Misc: 6 

Flight 

Flight Statistics: 

● Distance flown: 5.118 km 

● Time flown: 2.13 hrs 

● Number of Flights: 16 flights 

Flight Length (per day): 

July 7:

367 meters

areas 1, 6, 7, 8, 9, 11, 23

July 8:

1336 meters

areas 5, 10, 12, 13, 15

July 10: July 14:

1429 meters 998 meters

areas 2, 14, 16, 17, 18 areas 3, 4, 19, 20, 21, 22

Project Details, Videos, and Images 

Project Overview: 

● Project Name: Lincoln (Abraham) Park Trash Cleanup Project 

● Location: Lincoln (Abraham) Park 

● Date: April 24, 2024 - July 18, 2024 

● Objective: Provide data on the amount of trash around the Lincoln Park area. As an experimental project, we decided to monitor the lower portion of Lincoln Park, south of La Salle Drive. 

Project Images: 6 of 30

Encountered Issues and Solutions 

Software 

1. Determining the optimal height, speed, light conditions, and video size for the trash detection software a. To determine the optimal height, speed, lighting, and video sizes, we took different trashes and recorded 20 different flights and tested them into the software. Because parks are very large, footaging them takes a lot of storage. This method is the best way to determine the optimal settings for our videos, because videos with the least data usage and high trash detection determine optimal settings for recording footage. 

2. Pairing the flight log from the drone system to the video footage to find out the exact coordinates and position of the trashes and deciphering flight logs, as they are normally given in a .DAT format after export a. Currently, the .DAT files are unreadable and we are working on that issue. 

   1. For a temporary situation, we screen record the video footage and overlay that footage on the map. Then we compare the overlaid map and video recording side by side to determine the trashes, and when the drone flies over a trash, we pin that location on the map. This is currently being done manually, but we are working with our developers to overcome that issue and make this entire system automatic. 

3. Training the software to be able to detect trashes at different angles and incorporate a wide variety of trashes. To train the software, we needed to highlight the trash at every frame in a video and show the software what to look for when combing for trash in the parks.

   1. Currently, we are still in the development and testing phase in the drone. We are working on training the software by using the trashes found in park footage to help the software know what to look for. We need to continue monitoring parks with the Chicago Park Foundation to continue training and improving the trash detection software. 

Hardware 

1. Setting waypoints include many navigation issues. DJI Drones have a sensor that surrounds the drone, feeding information on potential obstacles from all angles. Waypoint flight relies on GPS coordinates and sensors to navigate the path without any collisions. The waypoint system needs a 5 foot margin for flight, which means that the drone will be 5 feet within the GPS waypoint, and the drone needs a 5 foot radius all around the drone, including the top and bottom. Using waypoints can be difficult because if an obstacle comes within 3 feet of the drone, the waypoint system will be suspended. This increases the radius of the drone to 8 feet, which can be difficult when the drone needs to navigate between trees. 

   1. One solution is to remove the sensor system, which means potential collisions with the waypoint system. This is not an optimal solution, as the drone has a 5 foot margin and can be damaged easily due to fall damage from hitting tree branches.

   2. Another solution is to change the settings in the obstacle avoidance from brake to avoid. This could result in the drone accidently colliding with an obstacle on its altered flight. This could also result in the drone being misdirected. 

   3. A last solution is to manually control the drone in areas where there might be issues with obstacles. Larger and clearer areas can be monitored by preset waypoint flight points and largely cluttered areas can be videoed by manually controlling the drone and videoing the area. Controlling the drone would require the user to be adept in drone flight and actively monitor and control the drone, reducing the ease and productivity of the system. 

Gratitude

Thank You: We would like to extend our heartfelt thanks to the Chicago Park Foundation for partnering with Eccodrone to monitor Lincoln Park. Their collaboration has been instrumental in enhancing the health and preservation of the Lincoln Park green space. We look forward to working with them in the future. 

General Overview: The drone monitoring system has proven effective in maintaining the safety and cleanliness of Lincoln Park. The real-time data collection and analysis have significantly improved the drone trash detection software and waypoint flight system. Continued collaboration and innovation are key to sustaining these advancements and exploring new opportunities for maintaining our parks. 

Sincerely, 

Inika Bansal 

Founder and CEO of EccoDrone