Actuator Fault State Transition Dataset for On-Orbit Servicing Robots

#State prediction #fault-conditioned modeling #dynamics consistency evaluation #safety boundary evaluation #On-orbit robot operations #actuator fault analysis #spacecraft safety verification #robot control evaluation

6340
0.3GB
JSON
CATL
MOBIUSI INC

Updated:2026-11-03

AI Analysis & Value Prop

This simulation dataset represents state, control action, and subsequent state changes in servicing robots when thrusters, reaction wheels, manipulator joints, or end effectors experience failures, lag, or output deviations. It covers multiple actuator fault conditions and provides observed and predicted states, transition residuals, dynamics-consistency checks, and safety-boundary verification signals. The data supports training and evaluating fault-conditioned state prediction models and reinforcement learning with verifiable rewards (RLVR).

Dataset Insights

Sample Examples

{}

Technical Specifications

Field	Type	Description
file_name	string	File name
action	object	Control command applied to the robot actuators in the current state.
current_state	object	Observed state of the servicing robot and its actuators before the fault-affected action is applied.
fault_condition	object	Type, location, and parameters of the actuator fault in the sample.
rlvr_evaluation	object	Reinforcement learning verification signals calculated from state prediction error, dynamics consistency, and safety-boundary results.
observed_next_state	object	Next-step state observed after applying the control action and accounting for the fault.
dynamics_consistency	object	Checks whether the state transition under the fault condition conforms to actuator dynamics and energy-change patterns.
predicted_next_state	object	Predicted next state calculated from the current state, control action, and fault condition.
state_transition_residual	object	Error statistics comparing the predicted next state with the observed next state.
safety_boundary_verification	object	Checks whether motion, output, temperature, and overall state remain within safety limits after the fault-affected transition.

Compliance Statement

Authorization Type	Proprietary - Commercial AI Training License (No Redistribution)
Commercial Use	Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization	No PII, no real company names, simulated scenarios follow industry standards
Compliance System	Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs

Frequently Asked Questions

Which actuator types and faults are covered?: The data covers thrusters, reaction wheels, manipulator joints, and end effectors, with simulated failures, lag, and output deviations. Each record links the fault condition and control action to the resulting state change.

What state transition verification data is included?: Along with the current, observed next, and predicted next states, the records include state transition residuals, dynamics-consistency checks under fault conditions, safety-boundary verification, and RLVR evaluation results. These fields help assess prediction error and constraint compliance.

Which training and evaluation tasks does it support?: The dataset supports fault-conditioned state prediction, state transition modeling, dynamics consistency evaluation, and safety boundary evaluation. It can also be used to build or assess reinforcement learning workflows based on verifiable reward signals.

How can I access the data, and what are the licensing terms?: Check the dataset product page for available access methods, delivery details, and licensing terms. Usage rights, redistribution restrictions, and commercial-use permissions are governed by the license published there.

Can't find the data you need?

Post a request and let data providers reach out to you.

Get this Dataset

Verified for Enterprise Use

Cite this Work

@dataset{Mobiusi2026,
  title={Actuator Fault State Transition Dataset for On-Orbit Servicing Robots},
  author={MOBIUSI INC},
  year={2026},
  url={https://www.mobiusi.com/datasets/94e465b5b31eee5fb9435de03b181a9f},
  urldate={2026-10-28},
  keywords={on-orbit servicing robot dataset,actuator fault data,state transition prediction,robot fault simulation,thruster failure,reaction wheel fault,RLVR training data,safety boundary verification},
  version={1.0}
}

Using this in research? Please cite us.

placeholder

Scene

Domain

Modality

Task