FDOT’s Bold Bet on AI Wildlife Detection: How a $0 Budget Ceiling RFP Could Save Florida Panthers on SR 40

FDOT District Five is seeking cutting-edge animal detection technology to protect drivers — and endangered species — along one of Florida’s most dangerous wildlife corridors. Here’s what vendors need to know.

The Problem: A Highway Through the Wild

If you’ve ever driven State Route 40 through the Ocala National Forest, you know the feeling. Dense tree canopy pressing in from both sides. A deer frozen in your headlights at dusk. Maybe a dark shape lumbering across the shoulder that you only recognized as a black bear after your heart rate came back down.

SR 40 cuts through one of the largest contiguous forested areas in Florida — home to deer, black bears, wild hogs, bobcats, coyotes, and yes, the critically endangered Florida panther. It’s a stunning stretch of road. It’s also a wildlife collision hotspot.

Now, FDOT District Five wants to do something about it — and the approach they’re taking is unlike anything we’ve seen in a Florida transportation procurement.

The Solution: AI Meets the Ocala National Forest

On April 6, 2026, FDOT District Five released RFP DOT-RFP-26-5003-WCAS — a Request for Proposal for a Wildlife Collision-Avoidance System (WCAS) to be deployed across eight previously installed ITS poles along SR 40 in Marion and Lake Counties.

The concept is straightforward, even if the technology behind it isn’t: deploy sensor arrays capable of detecting large animals approaching the roadway in real time, then automatically trigger flashing warning beacons to alert drivers before a collision occurs.

Think of it as a smart, automated version of those “Wildlife Crossing” signs — except these actually know when an animal is there.

What FDOT Is Looking For

Let me break down the key technical requirements, because this is where it gets interesting for vendors in the AI, computer vision, and sensor technology space.

Detection capabilities. The system must identify “macro species” — deer, bears, panthers, coyotes, bobcats, and wild hogs — while filtering out smaller animals like raccoons, opossums, and rabbits. The system isn’t required to detect animals that spend most of their lifecycle below the grass line, such as snakes and insects.

Accuracy thresholds. FDOT has set specific, measurable performance requirements:

• 80% true-positive accuracy or better in clear conditions
• 75% true-positive accuracy in adverse conditions (heavy rain, wildfire smoke, fog)
• Less than 20% false-positive rate
• Less than 5% false-negative rate
• 90% accuracy for species-level identification in reporting

That last bullet is notable — the system doesn’t need to distinguish between species to trigger a warning beacon, but it does need to identify specific species for data collection and reporting purposes. That’s a meaningful distinction for vendors designing their detection algorithms.

Speed. The system must generate a detection event and actuate the flashing warning beacon within one second of an animal entering the warning area. In a wildlife collision scenario, one second can mean everything.

Edge computing — no cloud allowed. Here’s where FDOT’s cybersecurity posture directly shapes the technical solution. All detection processing must happen at the roadside. Cloud-hosted solutions are explicitly prohibited for generating detection events, and the Department will not provide an outbound internet connection for the field equipment. This is consistent with FDOT’s increasingly security-conscious approach to ITS deployments and Executive Order 22-216 requirements.

Vendors can use a hybrid-cloud approach for reporting and data aggregation, but the core detection intelligence must live at the edge.

Environmental hardening. This system will live outdoors in Central Florida — year-round. The Ocala National Forest is also prone to significant wildfire and smoke conditions, which the RFP specifically calls out. Systems must comply with NEMA TS2 environmental standards and operate reliably through every weather and lighting condition Florida can throw at them.

The “Prove It First” Approach: A Smart Procurement Structure

What really stands out about this RFP is the phased task work order structure. FDOT isn’t just asking vendors to promise their system works — they’re requiring vendors to prove it before full deployment.

Task Work Order 1: Permits and BABA Compliance (up to 6 months) The vendor must secure a Traffic Control Device Permit and demonstrate Build America, Buy America Act compliance. This is the table-stakes phase — if you can’t clear these hurdles, you don’t move forward.

Task Work Order 2: Evaluation Period at the Wekiva Parkway Test Site (up to 12 months) This is the critical gate. The vendor must install a complete assembly at a controlled testing site along SR 429 (Wekiva Parkway) and demonstrate that the system actually works. FDOT will independently validate performance using trail cameras placed throughout the site, cross-referencing the vendor’s detection events against ground truth.

The minimum evaluation period is two weeks of data collection, and if no positive wildlife detections are confirmed by FDOT’s verification cameras, the evaluation can be extended. The vendor must pass a testing matrix with all criteria scored as PASS to achieve Final Acceptance.

During my time at FDOT, I saw too many technology deployments where the agency accepted vendor promises at face value. This “prove it in the field first” approach is exactly the kind of procurement discipline that protects both the agency and taxpayers.

Task Work Order 3: Full-Scale Deployment on SR 40 (remaining contract time) Only after successfully completing the evaluation period does the vendor proceed to deploy across all eight sites on SR 40. The vendor provides a turn-key solution and is responsible for everything — materials, installation, configuration, integration, testing, and one year of warranty support following Final Acceptance.

Strategic Considerations for Vendors

If you’re a technology company in the wildlife detection, computer vision, or roadside sensor space, here’s what you should be thinking about:

This is a niche market with growing demand. Wildlife-vehicle collisions are a nationwide problem, and transportation agencies across the country are exploring technology-based solutions. A successful deployment on SR 40 could be a powerful reference project.

The vendor pool is likely small. The combination of edge-computing AI, outdoor-rated sensor arrays, species-level identification, and integration with FDOT’s existing ITS infrastructure (SunGuide, managed Ethernet switches, ControlByWeb relays) creates a fairly narrow technical requirement. If your company has relevant capabilities, you may face limited competition.

BABA compliance matters. The Build America, Buy America Act requirements apply regardless of funding source — a point many vendors still miss. If your hardware is manufactured overseas, you need to start working through BABA compliance now, not after award.

The evaluation period is your audition. Your performance at the Wekiva Parkway test site will determine whether you get to deploy on SR 40. Bring your best system, your best team, and be prepared for FDOT to verify every detection event independently.

Think beyond the minimum. The RFP encourages vendors to identify potential value-added capabilities beyond the minimum technical requirements. If your system can do more — better species identification, predictive analytics, integration with connected vehicle technology — this is your chance to differentiate.

Scalability is mentioned explicitly. FDOT notes that “the number of sites is likely to expand in the future.” A successful vendor isn’t just winning eight sites on SR 40 — they’re potentially positioning themselves as FDOT’s go-to wildlife detection technology provider statewide.

Key Dates

• Advertisement Date: April 6, 2026
• Deadline for Questions: April 21, 2026 at 10:00 AM
• Technical Proposals Due: May 5, 2026 at 10:00 AM
• Price Proposal Opening: May 13, 2026 at 11:00 AM
• Intended Award Posting: May 18, 2026

All questions must be submitted in writing to [email protected]. Proposals must be submitted electronically to [email protected].

The Scoring: Technical Expertise Wins the Day

This is a best-value procurement weighted heavily toward technical capability:

• Technical Proposal: 100 points
  
  • Vendor Qualifications: 15 points
  • System Technical Approach: 60 points
  • Evaluation Period Plan: 25 points
• Price Proposal: 30 points (using the low-price formula)

With a 70-point minimum threshold on the technical proposal and price worth less than a quarter of the total evaluation, FDOT is sending a clear message: bring us the best technology, not just the cheapest bid. This is exactly the right approach for a novel technology deployment where system performance is literally a matter of life and death — for both drivers and endangered wildlife.

The Bottom Line

This procurement represents something genuinely exciting in Florida transportation technology. FDOT District Five is pushing the envelope by deploying AI-powered wildlife detection on a real-world corridor with real-world stakes — the Ocala National Forest, home to some of Florida’s most iconic and endangered species.

The phased approach protects the Department from vendor overpromises. The edge-computing requirement ensures cybersecurity. The detailed accuracy thresholds provide clear, measurable performance standards. And the emphasis on technical capability over price ensures that the winning system will be the one most likely to actually save lives.

For vendors with the right technology, this is more than a contract opportunity — it’s a chance to demonstrate that AI-powered wildlife collision avoidance works at scale on a Florida highway. And if it works on SR 40 through the Ocala National Forest, it can work anywhere.