PROTOTYPING SELF-HEALING BEHAVIOR FOR NASA SWARM-BASED SYSTEMS WITH ASSL

Abstract

Autonomic computing promises computer systems capable of self-management, which augurs great promise for unmanned spacecraft. Such spacecraft are extremely appropriate for deep space exploration missions because the former bring onboard intelligence and less reliance on control links. The Autonomic System Specification Language (ASSL) is a framework for developing autonomic systems. As part of our research on ASSL, we have successfully specified autonomic properties, verified their consistency, and generated implementation for both the NASA ANTS (Autonomous Nano-Technology Swarm) concept mission and the NASA Voyager mission. This paper presents concrete results on the use of ASSL to develop a self-healing behavior model for NASA ANTS swarm-based exploration missions. Here, we present specification and implementation results. Moreover, we experiment with the ASSL-generated code to demonstrate that the implemented ANTS system is capable of self-management in respect of the specified self-healing model.

Keywords

• intelligent swarms
• formal methods
• autonomic computing
• ASSL
• ANTS