Unitree Go1 Pro Repair Guide: Sensor, Gait and Power Evidence

Robotics repair branch

A Unitree Go1 Pro concern must stay tied to the exact Pro unit, its supplied kit, current state, event history and required task. The Go1 Pro name does not make every gait, sensor, power or connection symptom interchangeable with a Go1. This guide builds a model-specific evidence record, keeps unsafe owner-side experimentation out of the path and turns the customer's concern into transparent scope, approval and controlled acceptance.

Quick answer

Identify the Go1 Pro first, then separate sensor evidence from gait, power and connection behavior

Photograph the Go1 Pro label, front sensor face, blue carry handle, side ports, joint housings, legs, feet, battery, charger and controller before settings or parts change. Preserve the last normal task, event, environment and visible status. Stop for impact, swelling, liquid, binding, odor, abnormal heat or uncontrolled movement. Sensor evidence, gait evidence, power evidence and network evidence should be recorded as separate branches, then joined only where a controlled comparison supports the relationship.

What evidence should be captured before the customer acts?

How should common Unitree Go1 Pro symptoms be separated?

How do you prove the robot is a Unitree Go1 Pro before repair?

Start with the physical robot and photograph the model label, full body, front sensor face, blue carry handle, side ports, joint housings, leg structure, feet, battery area and every supplied accessory. Record the controller, charger and any added device separately. Do not use a listing title or a family resemblance to decide that the unit is a Go1 Pro, because evidence and supported configuration must remain tied to the exact robot on the bench.

Keep version information and the present working baseline in the same file. If several Unitree quadrupeds are stored together, label the evidence set before batteries, controllers or media are moved. The identity step is not administrative overhead: it determines which support information applies and prevents a known-good comparison from another model being mistaken for proof about the customer's Go1 Pro.

What should the Go1 Pro intake record capture?

Photograph all sides while the robot is powered off, then capture close views of the reported area, sensor windows, panel gaps, fasteners, joint housings, lower legs, foot pads, battery, ports and packaging. Preserve original incident video and the customer's wording. Record transport, fall, collision, moisture, dust, storage, battery substitution, controller change, prior service and software change in time order.

Define the required task before a quote is written. A customer may need stable indoor training, controlled research movement, remote observation or repeatable walking on a specific surface. Name the route, lighting, obstacles, control method, runtime and deadline. A result that powers on in a workshop can still fail the business or learning outcome the customer actually needs.

How should Go1 Pro sensor evidence be collected?

Record lighting, floor texture, reflections, obstacles, distance, dust, moisture and operating mode when perception behavior changes. Photograph sensor exteriors without touching or covering them, then capture the robot's response from a safe observer position. Repeat only a modest controlled scene after power and movement risks are screened. Do not build an artificial hazard to prove an avoidance function.

Sensor evidence should state what the robot saw or did under named conditions, not declare a hidden sensor defective. A changed response can involve the environment, surface, sensor exterior, configuration, communication or another system. Keep those possibilities open until a trained comparison narrows them. If the robot's movement is unstable, stop; collecting more perception footage does not justify unsafe locomotion.

Why is sensor behavior separate from connection behavior?

A robot can have a usable connection while responding unexpectedly to its environment, or it can lose app or controller access while its physical sensor hardware remains unchanged. Record discovery, connection, status visibility, command receipt, robot response and perception result as separate stages. Include device, network, distance and obstacles so the timeline can be interpreted later.

Do not treat reconnection as proof that every sensor and movement layer passed. Likewise, do not label a connection drop as a sensor failure simply because the customer could not see status. A transparent repair scope may contain two branches with separate known and unknown findings. That is more useful than one broad line item called electronics repair.

How do you document Go1 Pro gait and joint symptoms?

After structural, battery and uncontrolled-movement risks are bounded, use a level padded lane and record front, side and rear movement at a modest speed. Keep surface, mode, load, battery state and temperature stable. Note the phase, side and condition in which the difference appears. Preserve sound separately from speech and workshop noise so the record does not rely on memory.

Gait combines structure, joints, calibration, sensing, control, power and surface interaction. A short step or hesitation does not identify one actuator. Compare one condition at a time and stop for binding, abnormal heat, visible displacement or uncontrolled movement. The useful output is a repeatable pattern and a bounded diagnostic question, not a dramatic clip.

What can an exterior leg and foot inspection establish?

Use broad light to inspect joint shells, leg structure, fasteners, foot pads, panel gaps and visible contact marks. Record whether a gap, scuff or loose-looking area aligns with the reported side and event. Do not pry panels, force joints or tighten unknown fasteners as an owner-side test. Preserve the as-received position before any approved workshop action changes it.

Exterior evidence can support an impact or wear question but cannot prove the internal condition. A clean surface does not rule out a deeper issue, and a marked foot does not explain every gait symptom. The written file should separate observation from inference. That distinction lets the customer see why professional inspection is proposed and which unknowns remain.

How should Go1 Pro battery and shutdown evidence be recorded?

Identify the battery and charger, storage history, charge state, startup sequence, idle stability and exact task stage when power changes. Record whether the symptom appears during connection, standing, walking or sensor-heavy operation. Stop immediately for swelling, damaged casing, leakage, odor, abnormal heat or unstable power. Do not cycle an unsafe pack to collect a second result.

A charge indicator does not prove task-level battery health, and a movement shutdown does not prove the battery is the sole cause. A workshop comparison must name every substituted battery, charger or setting. Acceptance returns to the customer's supplied path or clearly states what remains outside scope. This prevents a known-good comparison from being presented as a universal cure.

What changes after impact, moisture or rough transport?

Preserve packaging, fall direction, contact surface and first observed stance. Photograph sensor face, body panels, joint housings, legs, feet, battery area and ports before repeated movement. Moisture, salt, impact near the battery, binding, odor, heat or uncontrolled motion moves the case out of owner-side powered testing. A dry exterior or a successful startup does not remove deeper uncertainty.

If carrier evidence or warranty scope may matter, keep as-received and post-test records separate. The customer should be able to see which marks existed at intake and which comparisons were later approved. Do not make the robot walk simply to prove it survived shipping. Physical screening and documented status come before controlled movement.

How do you compare Go1 Pro with Go1 without merging their diagnosis?

A Go1 can help frame a family-level question, but it is not an interchangeable baseline for every Go1 Pro. Record exact model, version, sensor configuration, supplied kit and task before any comparison. If a controller, battery or complete robot is used as a known-good reference, state precisely what changed and what the comparison can and cannot show.

A successful Go1 comparison does not prove that the Pro unit's full sensor, power or movement path is healthy. Likewise, a family difference is not automatically a fault. Keep model-specific evidence in the survivor page so searchers and customers reach the correct support path instead of a generic quadruped answer that overstates compatibility.

What belongs in a transparent Go1 Pro repair scope?

The scope should identify the exact unit, supplied kit, customer's concern, event timeline, reproduced behavior, known and unknown findings, diagnostic boundary and proposed work. It should state parts provenance, timing, acceptance plan and what would trigger a revised quote. It should not promise a board, joint, sensor or battery solely from a short symptom description.

Reboot Hub keeps evidence visible before payment and through handover. Every comparison, software change, part substitution and test condition is named. The customer can then decide whether the scope fits their task and tolerance for uncertainty. Transparent service means unknowns are stated early rather than converted into confident language after the invoice is paid.

When is a documented replacement stronger than Go1 Pro repair?

Repair is stronger when the symptom is safely repeatable, the model and baseline are understood, parts provenance is defensible and controlled acceptance can demonstrate the required task. Replacement is stronger when impact or battery risk cannot be bounded, multiple layers are affected, version history is uncertain, important parts are unavailable or the required outcome cannot be demonstrated within the customer's timeline.

A replacement comparison should name the exact Pro configuration, supplied equipment, visible condition, warranty boundary and residual unknowns. It should serve the customer's operation, not become an automatic upsell. Downtime, environment, control method, surface and evidence quality all belong in the decision beside cost.

What does controlled acceptance look like for Go1 Pro?

Controlled acceptance progresses through safe startup, status visibility, connection, stationary response, low-speed gait, sensor evidence and the customer's representative route. Record surface, lighting, obstacles, mode, load, battery, duration and repeatability. Stop if warnings, heat, sound, instability, perception anomalies or uncontrolled movement return. One clean power-on or one step is not enough.

Handover includes the original concern, written findings, approved work, parts provenance, before-and-after evidence, residual risks and operating boundaries. Reboot Hub's normal workshop target is 1-3 business days after quote approval when diagnosis, parts and acceptance allow. Approved work carries a 30-day repair warranty; eligible pre-owned complete products follow a separate 180-day product warranty. Those promises remain explicit and separate.

Related Reboot Hub paths

Move from evidence to a documented next step

Continue the learning path, submit the exact-unit concern or compare a documented replacement.