An Open and Modular Architecture

The ADVOCATE architecture is distributed around a CORBA bus. This is a guarantee of applicability whatever the module environment, language and location are.
The architecture is modular, easy to evolve and to adapt on existing systems. It comprises four different types of modules.

The ADVOCATE Modules
Vehicle Piloting Module (VPM)
This module manages the mission plans and communicates directly with the vehicle sensors and actuators. It is able to provide sensor data and actuator data to the other modules. Several VPMs, each of them working on a specific sub-system, can be plugged on the ADVOCATE architecture, all using the same communication interfaces.

Intelligent Fault Recovery System (IFRS) module
This module has a co-ordinating role, knowing the role of each (piloting and diagnosis) module, and managing the communications in between them. It is in charge of collecting alarms, requesting the diagnose results and the related recovery actions, converting them into mission plan modifications, and communicating these changes to the MMI and/or the VPM. It comprises a generic part, which is able to integrate different kinds of artificial intelligent techniques results (using different uncertainty measures), and a specific part that includes knowledge of the monitored vehicle.

Diagnosis module (DM)
Several diagnosis modules can be plugged onto the ADVOCATE architecture. Diagnosis Modules include modules providing a diagnosis (identification of sub-system state), a proposed recovery action, or both. These modules comprise a specific part based on vehicle knowledge, but they have all the same communication interface.

Man Machine Interface (MMI)
This module is the generic ADVOCATE operator interface. It displays alarms as well as results from other modules for validation. This module is not implemented in case of AUVs (no pilot).

Technologies of Artificial Intelligence Used
Bayesian Networks (BN)
The BN is a model representing the causal relations between the entities of the modelled domain.
An influence diagram adds decisions and value functions to the model. The strengths of the relations are described using probabilities. Utility functions describe the preferences of the decision-maker.
BNs and IDs can be adapted to many classification (diagnosis) or decision problems, particularly in case of erroneous, incomplete or uncertain data, or problems that involve sensitivity analysis, conflict analysis, or calculation of value of information.

Neuro-symbolic system (NSS)An incremental neuro-symbolic system (INSS) represents the initial expertise of the domain as symbolic rules written by the experts. These rules are compiled into a neural structure to be used during the on-line diagnosis. The compiled neural network is trained and tested on a set of representative examples. The refinement of the neural network can be performed when badly classified examples are encountered during the system functioning. These new examples are then added to the initial learning base. The knowledge of the system is then increased and the conservation of the initial knowledge is guaranteed.

Fuzzy Logic (FL)
A Fuzzy system represents (symbolic) expert knowledge by means of fuzzy rules. Fuzzy rules use linguistic variables (which values are linguistic labels) to describe a decision or control protocol in terms that are quite close to the language used by the experts. That "proximity" between the language used by the experts and that representing the fuzzy rules simplifies the process of knowledge extraction, and makes the decision process understandable by the experts. In addition, the underlying reasoning methods are particularly well adapted to decision or control problems working with uncertain or noisy data.

ADVOCATE Applications
Two end-users are involved in the ADVOCATE project:

• IFREMER (France) designing ROVs (Remotely Operated Vehicles) for scientific applications
• STN ATLAS Elektronik (Germany) designing AUVs (Autonomous Underwater Vehicles) and semi-AUVs for industrial applications

The ADVOCATE architecture and the communication protocol have been designed to be generic while still adapted to the two applications.

The developed modules for this architecture may contain vehicle-specific parts to fulfil the requirements of the two cases.

Future Evolutions and Perspectives
It has been proved during the project that is was possible to determine a common architecture and communication protocol for underwater vehicles coming from different manufacturers, and for intelligent modules based on different technologies.

The ADVOCATE project did not permit the implementation on real underwater vehicles to evaluate the results of operational use. That will constitute, in refining the models, the first next step in the exploitation of the project results.
It appeared also during the project that, although the architecture is generic, it will be of particular interest to provide a friendly tool to configure the specific parts of the modules.

Moreover, adaptation and application and tests of the ADVOCATE architecture and modules to other kinds of vehicles, or even industrial processes, will be also a valuable challenge to prove and extend the validity of the ADVOCATE concept.

You can find the official ADVOCATE Web site here .