Subscribe
All summer long, innovation has been the emphasis for the U.S. Army Corps of Engineers and its nine divisions and 45 districts spanning the globe.
Since the spring of 2024, the Kansas City District has been leading the way by helping pilot an innovative program — which combines GoPro cameras with an AI analysis tool to survey the conditions of road and parking lot surfaces across the district’s 18 lake projects.
“We attach the GoPro to the hood of a vehicle and simply drive the roads and parking lots while the GoPro takes geotagged photos,” said Kylie Ward, the asset manager for the Kansas City District Operations Division.
According to Ward, the camera snaps a picture at five second intervals, looking for hazards or defects. The photos are geotagged so they can be overlaid on a map to show what road segment and parking lot they belong to.
Then, that footage is provided to the contractor who built the AI tool specifically for USACE to flag and rate road surface deficiencies, placing them into a range of categories defined within the Recreational Operational Condition Assessment rating guidelines. Another positive is that the contracted portion of this process is funded by the Federal Lands Transportation Program.
USACE has used the Operational Condition Assessment tool for years to rate the conditions of roadways, parking lots and other recreational infrastructure maintained by USACE. The tool outlines common pavement defects such as alligator cracks, rutting or distortions, patches, road raveling and potholes just to name a few, that are calculated to give each road segment and parking lot a rating on a grade school style scale ranging from A for excellent, to F for failed.
Ward said the new technological approach to rating the hundreds of miles of roadways the Kansas City District owns and maintains, reduces time in the field compared with traditional walk-through inspections — while producing nationally consistent data that in theory, will help drive more accurate maintenance planning and funding prioritization in the future.
“By automating the survey and rating process, we reduce the subjectivity that sometimes comes with visual inspections,” she said.
Streamlining condition assessments
The data collected from each lake project feeds into USACE’s Operational Condition Assessment system, better known as OCA, which rates assets on a five-year cycle. Ward, who recently used this technology to survey Perry Lake in northwestern Kansas, said this new process saves a significant amount of time and man-hours.
“Normally road and parking lot condition assessments require two trained OCA team members plus one person from the project,” Ward said. “Utilizing the GoPro and AI technology allows it to be one person and a vehicle.”
Using this GoPro method, the person collecting the data doesn’t even have to be someone trained in OCAs. Once a district’s road and parking lot imagery is collected, it’s sent to Allison Borkowski, park ranger at the Mississippi River Project Office in the Rock Island District. Allison serves as the contracting officer’s representative for this effort and works closely with the contractor to get the results back for USACE.
So basically, Borkowski functions as the liaison between parties. She said the workflow has become faster as the model improves, and formatting “kinks” are resolved.
“Last time it took us about five or six months for the first original batch of data,” Borkowski said. “Now that all those kinks are worked out, we’re expecting it to take less time.”
She said the current task order is larger than the first, covering about 40 lake projects, but the artificial intelligence model has been “dialed in and trained even more.”
After districts ship back cameras, Borkowski moves the images from the SIM card onto a standalone computer, overlays them on existing GIS road and parking layers and forwards the package to the contractor.
“It goes in and it reads all of the different deficiencies on the road,” she said.
Borkowski added that the lighter field footprint is part of the appeal. Project offices can use whoever is available — from lake managers to seasonal staff to capture all the imagery.
“You can send one person off to do it on their own,” she said. “It’s a great project for summer interns to go and do. You don’t necessarily have to have your upper management or supervisors going out to collect the data.”
She also said that feedback has improved exponentially as more lake projects participate.
“Before people collect, they seem a little worried,” she said. “By the time people have been sending the cameras back, I’ve had good feedback. A lot of people say it’s a lot easier than they thought it was going to be.”
The team sends projects in batches so the contractor isn’t overwhelmed, and results can be compared in waves. Borkowski said the current contract is expected to run through five task orders.
Looking ahead
If performance remains strong and funding persists, the method could continue beyond the current agreement.
“It’s looking more successful with each task order,” Borkowski said. “We’re getting more streamlined, so it looks like this is the way it’s trending.”
Kylie Ward agreed and said the efficiency gains matter because they enable districts to focus on repairs that most affect visitors.
“The more data we collect, the better we can prioritize repairs and stretch taxpayer dollars further,” she said.
Borkowski said that USACE headquarters and the OCA teams meet monthly to review progress and funding. She also made it clear that she appreciates the coordination that allows money to move quickly from USACE headquarters to projects through the OCA program and its partners.
“It’s clear to see the trickle down of money that begins at headquarters and how it gets all the way down to the project in terms of the data collecting and eventually the data is uploaded into our OCA system,” she said.
As the next batch goes out, Borkowski said districts are eager to compare the second wave of results with the first in hopes to keep refining the process.
“We’re still in our beginning stages,” she said. “It’ll be interesting to compare our results.”
| Date Taken: | 08.25.2025 |
| Date Posted: | 08.25.2025 16:04 |
| Story ID: | 546417 |
| Location: | US |
| Web Views: | 17 |
| Downloads: | 0 |
This work, The Kansas City District tests AI and GoPro technology to improve roadway and campground maintenance with new pilot program, by Lawrence Brooks, identified by DVIDS, must comply with the restrictions shown on https://www.dvidshub.net/about/copyright.
