An AV-ready ground node is a private site upgraded with defined zones, access rules, service capability, a reservation workflow, SLA expectations, and evidence capture so an autonomous fleet can use it as trusted ground-service capacity.
An AV-ready ground node is an ordinary private site that has been structured so an autonomous fleet can reserve it, enter it, get work done there, and prove the session happened correctly. A plain lot has space. A node has space plus seven things a lot does not: defined zones, written access rules, declared service capability, a reservation workflow, check-in/check-out records, SLA expectations, and evidence capture. The fleet does not need more asphalt. It needs places where reservation, access, workflow, and proof are already solved.
| Capability | Plain private lot | AV-ready ground node |
|---|---|---|
| Physical space | Yes | Yes |
| Defined zones (stage, store, queue, hold) | No | Yes |
| Written access rules (who, when, how) | Informal | Explicit, machine-readable |
| Declared service capability | Unknown | Listed per node |
| Reservation and session record | None | Per visit |
| SLA expectations | None | Defined and tracked |
| Evidence capture (proof of session) | None | Captured and auditable |
What "AV-ready" actually means for a private site
AV-ready does not mean the site is mapped, autonomous, or certified for safety. It means the site has been qualified and structured so a fleet operator can treat it as dependable ground-service capacity. The distinction matters because mapping and readiness solve different problems. Mapping tells a vehicle where it may be able to navigate. Node readiness tells a fleet where the vehicle is allowed to go, what it is allowed to do there, whether the space and service are available, what rules apply, and what evidence proves the session ran correctly.
The property defines permission. The fleet validates capability. XoomPark coordinates reservation, access, workflow, SLA, evidence, and audit. A site becomes a node at the point where all six of those coordination layers exist for it, not when it merely has open spaces.
Why robotaxi fleets need nodes, not just lots
The driver disappears with autonomy. The ground work does not. Between trips a robotaxi still needs to stage, store overnight, queue near charging, get cleaned, get a visual inspection, get recovered after an incident, and reset. Those jobs used to ride along with a human driver. Now they are a distributed operations problem the fleet has to solve across a city.
Depots solve part of this, but a depot is a single fixed point. Waymo runs centralized depots where vehicles return to be charged, cleaned, and inspected by on-site technicians in its launch cities (Waymo; understandingai.org, 2025). A fleet running across a metro cannot send every idle vehicle back to one depot for every reset without burning deadhead miles, and roughly 40-44% of robotaxi vehicle-miles already run empty, in line with human ride-hail (Transport Findings, "Millions of Trips, Waymo Empty Miles," 2025). Distributed nodes let the work happen closer to where vehicles actually idle. That is the gap a node network fills: off-depot fleet support at points the fleet does not own.
How a private site becomes a ground node
The conversion is a workflow, not a construction project. Most of what makes a lot AV-ready is information and rules, not concrete. The sequence is consistent across site types:
| Step | What happens | Output |
|---|---|---|
| 1. Discovery | Site is identified as candidate capacity near fleet demand | Candidate record |
| 2. Qualification | Site is scored against the readiness framework | Readiness score |
| 3. Zone definition | Areas are designated for staging, storage, queueing, holding | Zone map |
| 4. Access rules | Owner declares who may enter, when, and under what conditions | Access policy |
| 5. Capability listing | Owner declares services available (cleaning, inspection space, power adjacency) | Capability sheet |
| 6. Reservation workflow | Fleet can request and confirm a session | Reservation record |
| 7. SLA + evidence | Expectations are set; check-in/out and proof are captured | Session record + audit trail |
Steps 1 and 2 are where a plain lot is separated from a usable node. The rest is structuring.
Who needs AV-ready ground nodes
Three groups have a direct stake. Fleet operators (the Waymo, Zoox, Nuro, and Avis Budget Group ground teams of the world) need distributed capacity they can trust without owning it. Property and parking operators (commercial lots, garages, underused industrial parcels) hold the capacity and want to monetize idle space without becoming AV experts. Fleet-ops and eMobility partners (Moove, Transdev, ABM, and charging-adjacent operators) sit in the middle and need a structured way to match supply to demand. Moove finances and operates Waymo's fleet in Phoenix and Miami, Transdev has run Waymo fleet operations since 2019, and ABM is a facility-services provider whose work intersects AV ground operations (TechCrunch, 2024; AVFleetTech, 2025).
Investors evaluating the category care about a fourth thing: that node readiness is a repeatable, scorable process rather than a bespoke negotiation per site. A framework that scores sites the same way every time is what makes a distributed network financeable.
The eight ground-service jobs a robotaxi node may support
We decomposed the work a robotaxi needs between trips into eight distinct ground-service jobs. This is the framework behind what a node can be qualified to support. Not every node supports all eight; readiness is scored per job.
- Staging: short-dwell positioning between trips.
- Overnight storage: long-dwell idle parking.
- Charging-adjacent queueing: holding vehicles in order near a charger the node does not own.
- PUDO access: structured pick-up / drop-off on private ground.
- Recovery holding: a place to put a vehicle after an incident or fault until it is collected.
- Light cleaning: interior/exterior reset space.
- Visual inspection: a defined spot for a walk-around and image capture.
- Service coordination: scheduling third-party work the node hosts but does not perform.
XoomPark coordinates reservation, access, workflow, SLA, and evidence for these jobs. It is not the charger owner, the maintenance provider, or the autonomy company.
Original research: the node-readiness scorecard
A plain lot and an AV-ready node look identical on satellite imagery. The difference is structural, and structure can be scored. We built a seven-dimension readiness framework, weighted toward the factors that most often block a fleet from actually using a site. This is a model we apply during qualification, not a published industry standard.
| Dimension | What it scores | Indicative weight |
|---|---|---|
| Zone clarity | Are stage / store / queue / hold areas defined and unambiguous? | 20% |
| Access rules | Are entry conditions explicit, time-bound, and machine-readable? | 20% |
| Service capability | Which of the eight ground-service jobs can this node support? | 15% |
| Reservation readiness | Can a session be requested, confirmed, and recorded? | 15% |
| SLA definability | Can clear, trackable expectations be set for a session? | 10% |
| Evidence capture | Can check-in/out and proof-of-session be captured and audited? | 10% |
| Demand proximity | Is the site near where fleet vehicles actually idle? | 10% |
Methodology note: the weights above are illustrative of how we prioritize during qualification, not fixed coefficients, and would be calibrated against real qualification runs as pilot data accrues. As a planning assumption in this model, the single most common blocker is access ambiguity, not lack of space, which is why access rules and zone clarity carry the highest weight. A site can have abundant capacity and still score low if a fleet cannot get a clear, repeatable answer to "are we allowed in, and what proves we were here."
The practical output is a score plus a punch list: a lot scoring 40% gets a specific list of what would move it to 80%, usually rules and records rather than construction.
What XoomPark does and does not do at a node
| XoomPark does | XoomPark does not |
|---|---|
| Discover and qualify candidate sites | Own or operate chargers |
| Score node readiness | Run an EV charging network |
| Structure access rules and zones | Perform vehicle maintenance or cleaning |
| Run reservation and session records | Map roads or build HD maps |
| Track SLA and capture evidence | Dispatch vehicles or run autonomy |
| Produce billing and audit records | Replace fleet operators (Moove, Avis, Transdev, ABM, Waymo ops) |
Not for you if
If you run a single fixed depot and never operate off-site, you do not need a distributed node network yet. A depot you own, with your own staff and your own rules, already gives you access, workflow, and evidence in one place. Node readiness matters when vehicles idle away from the depot and you need trusted ground-service capacity at sites you do not control. If every reset already happens at home base, this is not your problem to solve today.
Frequently asked questions
What is an AV-ready ground node?
An AV-ready ground node is a private site structured so an autonomous fleet can reserve it, access it under clear rules, get defined ground-service work done there, and capture proof the session happened. It is the difference between a lot that has space and a site a fleet can actually depend on.
How do private sites become AV-ready?
Through qualification and structuring, not construction. A site is scored against a readiness framework, then given defined zones, written access rules, a capability listing, a reservation workflow, SLA expectations, and evidence capture. Most of the work is information and rules rather than concrete.
Are AV-ready ground nodes the same as parking?
No. Parking is storing a vehicle in a space. A ground node adds access rules, declared service capability, reservation and session records, SLA tracking, and evidence capture so a fleet can trust and audit what happened. Storage is one of eight jobs a node may support.
Why are depots not enough for robotaxi fleets?
A depot is a single fixed point, and routing every idle vehicle back to one depot for every reset burns deadhead miles across a metro, where roughly 40-44% of robotaxi miles already run empty (Transport Findings, 2025). Distributed nodes put staging, queueing, cleaning, and recovery closer to where vehicles actually idle.
Related pages
Request an AV-ready site scorecard
If you own a lot, garage, or parcel and want to know whether it could be a ground node, request an AV-ready site scorecard. You get a readiness score across all seven dimensions and a specific punch list of what would move it from a plain lot to usable fleet capacity.