Proceedings of the ASME 2020 International Mechanical Engineering Congress and Exposition IMECE2020 November 15-19, 2020, Portland, OR, USA IMECE2020-23234 FORMALIZING PERFORMANCE EVALUATION OF MOBILE MANIPULATOR ROBOTS USING CTML Omar Aboul-Enein ∗ Salisbury University Salisbury, Maryland National Institute of Standards and Technology Gaithersburg, Maryland Email: omar.aboul-enein@nist.gov Yaping Jing Salisbury University Salisbury, Maryland Roger Bostelman National Institute of Standards and Technology Gaithersburg, Maryland ABSTRACT Computation Tree Measurement Language (CTML) is a newly developed formal language that offers simultaneous model verification and performance evaluation measures. While the theory behind CTML has been established, the language has yet to be tested on a practical example. In this work, we wish to demonstrate the utility of CTML when applied to a real-world application based in manufacturing. Mobile manipulators may enable more flexible, dynamic workflows within industry. There- fore, an artifact-based performance measurement test method for mobile manipulator robots developed at the National Institute of Standards and Technology was selected for evaluation. Contri- butions of this work include the modeling of robot tasks imple- mented for the performance measurement test using Petri nets, as well as the formulation and execution of sample queries using CTML. To compare the numerical results, query support, ease of implementation, and empirical runtime of CTML to other tem- poral logics in such applications, the queries were re-formulated and evaluated using the PRISM Model Checker. Finally, a dis- cussion is included that considers future extensions of this work, relative to other existing research, that could potentially en- able the integration of CTML with Systems Modeling Language (SysML) and Product Life-cycle Management (PLM) software solutions. ∗ Address all correspondence to this author. 1 INTRODUCTION In the field of formal verification, Computation Tree Mea- surement Language (CTML) offers many advantages over con- ventional methods for conducting performance-reliability analy- sis. CTML combines performance and reliability evaluation ca- pabilities under one language [1]. Jing and Miner establish the theoretical foundation for CTML, describe the advantages of the language, and provide an example of the language usage through modeling the Dining Philosophers problem. In their work, a few primary advantages of CTML are noted. First, CTML is able to respond to nested queries with either a real-valued quantity or a probability. In addition, CTML matches the functionality of Probabilistic Computation Tree Logic (PCTL) [2] and can re- spond to a (non-trivial) subset of Probabilistic Linear Temporal Logic (PLTL) queries that are not expressible in PCTL [1, 3]. Most importantly, the functionality of CTML extends beyond the aforementioned logics as the language can answer queries not expressible in either PCTL or PLTL [1]. For example, CTML can answer survivability queries, which ask how much time re- mains until an event occurs given that another event has already occurred. Finally, Jing and Miner theoretically established the running efficiency of CTML as polynomial with respect to the both the formula and state size [1]. The theoretical advantages of CTML just described also sug- gest potential benefit towards Industry 4.0 applications. Indus- try 4.0 is the convergence of “robotics, cyber-physical systems, 1 Copyright c  2020 by ASME