Abstract
Robots are improving their autonomy with minimal human supervision. However, auditable actions, transparent decision processes, and new human-robot interaction models are still missing requirements to achieve extended robot autonomy. To tackle these challenges, we propose RODEO (RObotic DEcentralized Organization), a blockchain-based framework that integrates trust and accountability mechanisms for robots. This paper formalizes Decentralized Autonomous Organizations (DAOs) for service robots. First, it provides a ROS–ETH bridge between the DAO and the robots. Second, it offers templates that enable organizations (e.g., companies, universities) to integrate service robots into their operations. Third, it provides proof-verification mechanisms that allow robot actions to be auditable. In our experimental setup, a mobile robot was deployed as a trash collector in a lab scenario. The robot collects trash and uses a smart bin to sort and dispose of it correctly. Then, the robot submits a proof of the successful operation and is compensated in DAO tokens. Finally, the robot re-invests the acquired funds to purchase battery charging services. Data collected in a three day experiment show that the robot doubled its income and reinvested funds to extend its operating time. The proof-validation times of approximately one minute ensured verifiable task execution, while the accumulated robot income successfully funded up to 88 hours of future autonomous operation. The results of this research give insights about how robots and organizations can coordinate tasks and payments with auditable execution proofs and on-chain settlement.
Why this matters
Current service robots face three challenges that limit their integration into organizations:
Coordination
Organizations deploy robots from different vendors, each specialized in specific tasks like cleaning, delivery, or security. These deployments rely on proprietary management systems that, forcing organizations to manage heterogeneus robotic systems.
Expression
Robots cannot independently advertise what services they can perform or prove that a task was successfully completed. This inability to express capabilities and verify actions creates operational bottlenecks and prevents flexible task allocation across mixed robot fleets.
Trust
When a robot fails or causes damage, operational logs are typically stored on proprietary servers or the robot’s local disk—records that can be altered or deleted. This makes establishing trust and liability challenging, as the operational source of truth remains obscure and mutable.
RODEO addresses these challenges by providing a blockchain-based coordination layer where robots can register services, accept task assignments, submit verifiable execution proofs, and receive token-based compensation - all recorded on an immutable, decentralized ledger that ensures accountability and enables autonomous economic participation.
RODEO overview
RODEO is organized into three primary architectural building blocks that work together to enable autonomous, accountable robot operations within decentralized organizations:
DAO bridge
The DAO bridge serves as the interface layer between all participants in the blockchain network. It provides organizations, humans, and robots with a common programmable API to publish tasks and services, receive assignments, manage rewards, and observe state updates.
Blockchain network
The blockchain network addresses the critical issues of trust and liability. It functions as a transparent, tamper-proof ledger that records participant interactions, task completion proofs, and robot data in a way that cannot be altered. Smart contracts define organizational rules.
Verification oracle
The verification oracle provides an auditable source of truth by ensuring reward settlement is based on verified physical performance. It validates that task execution proofs are legitimate and unmanipulated and performs emulation-based verification.
Use Case: Cleaning and charging in a university lab
To evaluate RODEO, we designed a cleaning task scenario in a university lab. The main participants were an organization (managing tasks and services) and a mobile robot with a manipulator. Two services were established: waste disposal (provided by the robot) and battery charging (provided by the organization). In this scenario, when trash is found in the lab (normally left by the students), the organization creates a waste disposal task, and the system assigns it to an available robot.
The waste disposal workflow
- 1
Organization creates task
When trash is found in the lab, the organization creates a waste disposal task using a web interface. The system assigns it to an available robot.
- 2
Robot picks trash
The robot navigates to the tagged waste item, uses its manipulator to grasp the object, then drives to the iTrash smart bin.
- 3
Disposes correctly
The robot presents the grasped item to the iTrash camera, queries the classification service, and receives a completion flag after dropping the item in the correct bin.
- 4
Submits proof
After task completion, the robot uploads a rosbag file containing its base trajectory, manipulator poses, gripper release confirmation, and the iTrash success message as verifiable proof.
- 5
Oracle validates
The verification oracle downloads the rosbag, replays it in Gazebo at 3× speed, and verifies execution checkpoints: pick-and-place stops, manipulator poses, gripper release, and iTrash message.
- 6
Robot receives 100 IEC
Upon successful validation, the smart contract releases the escrowed 100 IEC tokens to the robot’s wallet, compensating it for the completed service.
The charging workflow: Reinvesting earnings
- 1
Battery low, robot creates charging task
When the robot’s battery reaches 50%, it autonomously creates a charging task and navigates to the charging area.
- 2
Charging costs 200 IEC
The robot completes a battery charging session at a fixed electricity tariff of 200 IEC paid to the organization.
- 3
Organization submits charging proof
After the charging session completes, the organization submits energy consumption logs as proof of service delivery.
- 4
Robot reinvests earned funds to extend autonomy
The oracle validates the charging proof, the smart contract deducts 200 IEC from the robot’s wallet, and the robot is ready to continue autonomous operation—effectively reinvesting its task earnings to extend its operational lifetime.
Real-world experiments
We ran our cleaning task scenario for three consecutive days in the AI & Robotics lab at IE University. During the experimental campaign we completed a total of 59 tasks where 8 were battery charging tasks created by the robot and 51 were waste disposal tasks created by the organization.
The prototype of RODEO was deployed on the Ethereum Sepolia Testnet. We deployed an ERC20 compatible token named IEC that was used for rewards in our experimentation. We deployed the following smart contracts: Organization.sol, TaskManager.sol, ServiceManager.sol, and IECoin.sol.
Results
Task completion and proof validation times
Cleaning tasks completed in about 4.1 minutes total, with oracle validation around 1.1 minutes. Across the three evaluation days, cleaning task execution time was 3.2 minutes (range: 2.4–3.7 minutes), covering navigation to trash, grasping the item, depositing it in the bin, and uploading the proof. Oracle decision time was 1.1 minutes (range: 0.9–1.2 minutes), reflecting rosbag replay at 3× speed while verifying execution checkpoints. Total cleaning task completion time (execution + validation) averaged 4.1 minutes (range: 3.6–4.8 minutes).
Charging took much longer, but proof verification stayed under about a minute. As expected, battery charging dominated overall runtime with a median of 65.5 minutes (range: 53.6–73.0 minutes across the three days). The robot took 65.5 minutes to fully charge the battery. However, the corresponding oracle verification time remained fast at 0.8, 1.0, and 0.3 minutes, yielding a median charging-task completion time of 65.8 minutes. The brief verification time demonstrates that proof validation scales independently of physical task duration.
Robot wallet balance evolution
Over three days, the robot’s balance increased from 2000 to 4100 IEC. The organization seeded the robot wallet with 2000 IEC at the start. Through autonomous task execution in the lab, the balance increased to 4100 IEC—meaning the robot could return the initial stake and still retain an additional 2100 IEC to sustain further operation. Drops in the yellow line mark charging events (red diamonds), while flat segments indicate inactivity during nighttime hours when the lab was unoccupied. The steady upward trend demonstrates that task earnings exceeded operational costs, validating the economic sustainability of the robot-driven DAO model.
Extended autonomy through reinvestment
Task earnings could fund 11 more charges, equal to 88 hours of operation. Daily task earnings, when reinvested as battery charges, enable extended autonomous operation. On day one, the robot completed 5 tasks (+500 IEC) and charged once (−200 IEC), resulting in a 300 IEC balance—enough to fund 1 additional charge and 8 hours of operation. Funded hours increased to 24 hours on day two and 56 hours on day three, corresponding to 3 and 7 charging cycles respectively. Across the three days, accumulated earnings could sustain the robot for 88 hours of future operation (11 total charges), demonstrating that task-based income successfully covers operational costs and enables continued autonomy without external funding.
Future work
There are several interesting extensions that could enhance RODEO’s real-world viability. First, expanding the experimental setup to include multiple robots and human participants would allow us to test whether token-based incentives measurably improve human-robot collaboration. For example, robots could post their own maintenance tasks (e.g., “clean my wheels” or “upgrade my hardware”), incentivizing humans to handle them carefully since their continued operation sustains future revenue. Additionally, large-scale simulations using Gazebo and PyRoboSim could help model the organization’s evolution under diverse conditions—varying robot populations, task availability, and pricing mechanisms—to identify economic policies that maintain sustainability as complexity grows.
From a technical perspective, scaling to multi-robot organizations will require addressing blockchain scalability challenges. As the number of robots and tasks increases, fluctuating gas fees and block confirmation delays could affect real-time responsiveness. Future work should explore Layer-2 scaling solutions or sidechain integrations to reduce costs and latency while preserving auditability. Finally, our current rosbag-based proof mechanism remains vulnerable to spoofed messages or replayed telemetry. Implementing hardware-backed signing or tamper-evident “sealed rosbags” would strengthen proof trustworthiness, ensuring that execution data submitted to the DAO has not been modified before validation.
BibTeX citation
@inproceedings{groshev2026rodeo, title={RODEO: RObotic DEcentralized Organization}, author={Groshev, Milan and Castell{\'o} Ferrer, Eduardo}, booktitle={IEEE International Conference on Robotics and Automation (ICRA)}, year={2026}, organization={IEEE}}