NOIR: Neural Signal Operated Intelligent Robots for Everyday Activities: Appendix 4

Authors: (1) Ruohan Zhang, Department of Computer Science, Stanford University, Institute for Human-Centered AI (HAI), Stanford University & Equally contributed; zharu@stanford.edu; (2) Sharon Lee, Department of Computer Science, Stanford University & Equally contributed; zharu@stanford.edu; (3) Minjune Hwang, Department of Computer Science, Stanford University & Equally contributed; zharu@stanford.edu; (4) Ayano Hiranaka, Department of Mechanical Engineering, Stanford University & Equally contributed; zharu@stanford.edu; (5) Chen Wang, Department of Computer Science, Stanford University; (6) Wensi Ai, Department of Computer Science, Stanford University; (7) Jin Jie Ryan Tan, Department of Computer Science, Stanford University; (8) Shreya Gupta, Department of Computer Science, Stanford University; (9) Yilun Hao, Department of Computer Science, Stanford University; (10) Ruohan Gao, Department of Computer Science, Stanford University; (11) Anthony Norcia, Department of Psychology, Stanford University (12) Li Fei-Fei, 1Department of Computer Science, Stanford University & Institute for Human-Centered AI (HAI), Stanford University; (13) Jiajun Wu, Department of Computer Science, Stanford University & Institute for Human-Centered AI (HAI), Stanford University. Table of Links Abstract & Introduction Brain-Robot Interface (BRI): Background The NOIR System Experiments Results Conclusion, Limitations, and Ethical Concerns Acknowledgments & References Appendix 1: Questions and Answers about NOIR Appendix 2: Comparison between Different Brain Recording Devices Appendix 3: System Setup Appendix 4: Task Definitions Appendix 5: Experimental Procedure Appendix 6: Decoding Algorithms Details Appendix 7: Robot Learning Algorithm Details Appendix 4: Task Definitions For systematic evaluation of task success, we provide formal definitions of our tasks in the format of BEHAVIOR Domain Definition Language (BDDL) language [69, 71]. BDDL is a predicate logic-based language that establishes a symbolic state representation built on predefined, meaningful predicates grounded in physical states [71]. Each task is defined in BDDL as an initial and goal condition parametrizing sets of possible initial states and satisfactory goal states, as shown in the figures at the end of the appendix. Compared to scene- or pose-specific definitions which are too restricted, BDDL is more intuitive to humans while providing concrete evaluation metrics for measuring task success. This paper is under CC 4.0 license. available on arxiv