U.S. DOT FY 2026 Phase I Pre-Solicitation Q&A
- To create a new account or to sign in with an existing account please click on “sign in” located at the top right of the page
- A new pop-up window will open to make a selection, you can use an existing Google or Facebook account to access the site or you might select to create a new account.
- If you choose to access UserVoice with an existing Google or Microsoft account, you will be required to enter your login credentials for Google or Microsoft and agree to terms of service
- If you need to create a new account, click "Create an Account" underneath the password box
- In the next screen, type in the e-mail and click "Verify E-mail".
- Navigate to your inbox and click the link within the e-mail to verify. The link will be generated by noreply@trymagic.com
- That link will be live for 20 minutes. Once that time elapses, you will have to request another one
- Navigate back to UserVoice, and you will need to provide your name, a password, and agree to terms of service and storage permissions. Click "Create Account".
- Select the research topic your question is related to from the drop-down menu that appears when you begin typing.
- You may add more detail in the Description box (optional). Do NOT attach any files. The U.S. DOT shall not consider any submitted materials other than questions.
- Enter your e-mail address and password.
- When finished, select “Post idea.” If you registered, you will be able to track responses to your post. Please include only one question or comment per post to help us keep the forum organized.
229 results found
-
Temporary work-zone congestion
Does topic 26-FH1 consider temporary work-zone congestion caused by roadway maintenance or robotic maintenance operations to be an acceptable Phase I use case for demonstrating Edge-AI/V2X congestion prediction and mitigation?
1 vote -
mobile autonomous aerial sensing platform
Would a mobile autonomous aerial sensing platform be considered an acceptable emerging data source if the proposed system’s main deliverable is an Edge-AI/V2X congestion advisory architecture that detects work-zone or incident-related congestion and delivers location-specific advisories to affected road users and traffic management operators?
1 vote -
For the Desired Outcome “Commercial and Implementation Readiness Assessment"
For the Desired Outcome “Commercial and Implementation Readiness Assessment”, what constitutes the minimum acceptable evidence of a commercialisation pathway — a Letter of Interest from a single state DOT or logistics partner, or are commitments from multiple stakeholders expected at Phase I?
1 vote -
For Technical Parameter TP-4 (federated learning)
For Technical Parameter TP-4 (federated learning): does Phase I require a live multi-party FL deployment involving distinct legal entities sharing model updates, or is a simulated federation — two data partitions on the same infrastructure representing separate jurisdictions — sufficient as a Phase I architecture demonstration?
1 vote -
Technical Parameter TP-1 requires deploying “low-power edge AI devices at strategic freight bottlenecks.
Technical Parameter TP-1 requires deploying “low-power edge AI devices at strategic freight bottlenecks.” For Phase I, is a benchtop hardware prototype or hardware-in-the-loop simulation of the edge device sufficient to demonstrate technical feasibility, or does Phase I require actual field deployment at a freight interchange?
1 vote -
For Technical Parameter TP-3 (synthetic data generation), does the DOT expect the generator to produce freight-domain-specific
For Technical Parameter TP-3 (synthetic data generation), does the DOT expect the generator to produce freight-domain-specific synthetic trajectories and delay events, or is general-purpose tabular synthetic data augmentation acceptable as a Phase I implementation of this capability?
1 vote -
26-FT1, Complete Trip Scope
For Topic 26-FT1, should the Complete Trip scope include pedestrian network segments before, between, and after transit use, including sidewalks, crossings, curb ramps, station access, stop access, and transfer paths?
1 vote -
1 vote
-
For the Freight Corridor Predictive Intelligence topic, is preference given to solutions that provide dual functionality — simultaneously mo
For the Freight Corridor Predictive Intelligence topic, is preference given to solutions that provide dual functionality — simultaneously monitoring both commercial vehicle health and the structural integrity of the weigh station platform itself using the same sensor array?
1 vote -
Is IRB-approved human-subjects testing required, or is lighter-touch evaluation ok?
For the human factors testing, is the program expecting IRB-approved human-subjects testing, or is a lighter-touch evaluation (heuristic review or expert walk-through) acceptable?
1 vote -
Must multi-agent coordination be demonstrated, or is single-agent with extensibility enough?
On the multi-agent coordination requirement, how much of that needs to be shown at this stage? A small demo with two adjacent edge nodes coordinating is a different scope than a single-agent demo with multi-agent extensibility just documented, so we'd like to scope it right.
1 vote -
Preferred archived data sources, or is sourcing offeror-driven?
For archived data, is there a preferred source (NPMRDS, RITIS, state DOT data lakes, the Work Zone Data Exchange), or is that really up to the offeror?
1 vote -
What CV *********** rate should the benefits analysis assume?
What connected-vehicle *********** rate should the benefits analysis assume? The topic asks for a with-vs-without-advisory comparison, and the result swings a lot depending on whether we're modeling at 5%, 10%, or 20% ***********.
1 vote -
Is CARMA / Saxton Lab integration expected for the HIL demo?
For the HIL setup, does the program want to see integration with CARMA / the Saxton Lab, or is it fine to use our own HIL environment? Trying to figure out if CARMA integration is something that actually matters competitively here.
1 vote -
C-V2X as baseline given 5.9 GHz reallocation, or is DSRC still acceptable?
With the 5.9 GHz reallocation and the move to C-V2X, should we be treating C-V2X (PC5/Uu) as the baseline for the HIL demo, or is DSRC still on the table?
1 vote -
Data integration plan — strategy document or working ingest pipelines?
Is this more of a strategy document, or do you want to see working ingest pipelines for at least some of the data sources by the end?
1 vote -
Digital twin expectation — microsimulation, generative AI, or existing platform integration?
For the digital twin piece — are you thinking more along the lines of physics-based microsimulation (Aimsun, SUMO, Vissim), or a generative AI model trained on real corridor traces? And is the program open to building on top of an existing digital-twin platform, or would you rather see something built from scratch?
1 vote -
Priority among data sources, and are licensed feeds (NPMRDS, ATRI) government-furnished?
Is there a priority order on the data sources? We're weighing NPMRDS, ATRI, state WIM, FAF, TPIMS, and connected vehicle BSMs, and it'd help to know which ones the program sees as most important. Also, for the licensed ones like NPMRDS and ATRI — are those typically furnished to awardees, or should we plan to budget for them?
1 vote -
Clarification on 80% accuracy target — spatial unit and error metric?
Quick clarification on the 80% accuracy target for 30–60 minute forecasts — is that measured at the segment level, at specific bottlenecks, or across full origin-destination pairs? And what error metric are you using? Whether it's MAPE on travel time vs. classification accuracy on a congestion label changes the modeling approach quite a bit.
1 vote -
Alternate approach to traffic congestion prevention and mitigation
This SBIR solicitation focuses on traffic congestion prevention and mitigation via advisories intended to optimize overall traffic flow. Would SBIR consider funding an alternate approach to traffic congestion prevention and mitigation, which allows all motor vehicles of any type to operate safely in near bumper-to-bumper proximity, thereby significantly increasing the throughput of all roadways? This is achievable without self-driving and with complete ad hock autonomy of each vehicle (i.e., no platoons or preplanning of any type). Fully automated acceleration-deceleration is achievable with adjacent lead-follow vehicle proximity dependencies including speed, braking performance, road conditions, and accommodations to permit ad hock vehicle…
1 vote