Purpose
Interpret roaming behaviour and separate network-layer issues from application issues.
Prerequisites
- Event timeline of AP association
- Application connectivity timeline
Safety considerations
- Follow your site's lockout / isolation and robot-zone entry rules before approaching equipment.
- Only trained, authorized personnel perform service actions. This page is concept training.
Normal behavior
The client reassociates to a better AP quickly; the application session continues over the IP network.
Conceptual operation
- Current association → candidate AP evaluation → reassociation → new association.
- Applications (APIs, MQTT) run above the network; they don't 'roam' themselves.
Inspection points
- AP coverage overlap
- Client roaming configuration (authorized)
Diagnostics
- Time between disassociation and reassociation
- Packet loss around the roaming event
- Application reconnect events
Common symptoms
- Brief offline periods at the same place
- Robot 'sticky' to a distant AP
Recovery decision tree
1. Does the app drop exactly at the roaming event?
YES → Investigate roaming duration/overlap.
NO → Look above the network layer (service, broker, certificates).
Authorized procedure
INTERNAL / AUTHORIZED ONLY
Roaming thresholds and client configuration values — available only in the model-specific, authorized SOP for your site. Not published on this website.
Verification
- Smooth roaming over repeated runs without application drops
Escalation
Network team with event timeline.