Detection of Motion Patterns and Transition Conditions for Automatic Flow Diagram Generation of Robotic Tasks

In this paper a method for detecting motion patterns and transitions between them in source code describing robotics tasks is proposed, being used to automatically generate flow diagrams of the corresponding task. This is done through the combination of both static and dynamic program analysis, first segmenting original code based on token matching and then using runtime data to judge importance of each segment and relations between them. Generation of flow diagrams is not only a simple way of visualizing overall procedure of complex tasks, but can also be used for current state identification and fail recovery systems. Proposed system is verified through experimentation using PR2 robot on the household task of cleaning up a table, being able to automatically generate a flow diagram with reasonable number of states.

[1]  Marvin Minsky,et al.  Computation : finite and infinite machines , 2016 .

[2]  Corrado Böhm,et al.  Flow diagrams, turing machines and languages with only two formation rules , 1966, CACM.

[3]  Edmund M. Clarke,et al.  Representing circuits more efficiently in symbolic model checking , 1991, 28th ACM/IEEE Design Automation Conference.

[4]  Kwang-Ting Cheng,et al.  Automatic Functional Test Generation Using The Extended Finite State Machine Model , 1993, 30th ACM/IEEE Design Automation Conference.

[5]  Paul G. Sorenson,et al.  Towards the Automatic Generation of Software Diagrams , 1991, IEEE Trans. Software Eng..

[6]  Robert S. Boyer,et al.  A Mechanical Proof of the Unsolvability of the Halting Problem , 1984, JACM.

[7]  Ahmad Naebi,et al.  A New Flowchart and Programming Technique Using Bond Graph for Mechatronic Systems , 2011, 2011 UkSim 13th International Conference on Computer Modelling and Simulation.

[8]  Dan Gusfield,et al.  Algorithms on Strings, Trees, and Sequences - Computer Science and Computational Biology , 1997 .

[9]  Edmund M. Clarke,et al.  Sequential circuit verification using symbolic model checking , 1991, DAC '90.

[10]  Kwang-Ting Cheng,et al.  Automatic generation of functional vectors using the extended finite state machine model , 1996, TODE.

[11]  William Landi,et al.  Undecidability of static analysis , 1992, LOPL.

[12]  Masayuki Inaba,et al.  Self-improving Robot Action Management System with Probabilistic Graphical Model Based on Task Related Memories , 2016, IAS.

[13]  Genny Tortora,et al.  Automatic Generation of Visual Programming Environments , 1995, Computer.

[14]  Randal E. Bryant,et al.  Graph-Based Algorithms for Boolean Function Manipulation , 1986, IEEE Transactions on Computers.