Collaborative Planning with Encoding of Users' High-Level Strategies

The generation of near-optimal plans for multi-agent systems with numerical states and temporal actions is computationally challenging. Current off-the-shelf planners can take a very long time before generating a near-optimal solution. In an effort to reduce plan computation time, increase the quality of the resulting plans, and make them more interpretable by humans, we explore collaborative planning techniques that actively involve human users in plan generation. Specifically, we explore a framework in which users provide high-level strategies encoded as soft preferences to guide the low-level search of the planner. Through human subject experimentation, we empirically demonstrate that this approach results in statistically significant improvements to plan quality, without substantially increasing computation time. We also show that the resulting plans achieve greater similarity to those generated by humans with regard to the produced sequences of actions, as compared to plans that do not incorporate userprovided strategies.

[1]  Jonathan P. How,et al.  Operator Object Function Guidance for a Real-Time Unmanned Vehicle Scheduling Algorithm , 2012, J. Aerosp. Comput. Inf. Commun..

[2]  Peter A. Jarvis,et al.  PASSAT: A User-centric Planning Framework , 2002 .

[3]  Dana S. Nau,et al.  SHOP2: An HTN Planning System , 2003, J. Artif. Intell. Res..

[4]  Jussi Rintanen,et al.  Complexity of Concurrent Temporal Planning , 2007, ICAPS.

[5]  Carmel Domshlak,et al.  Reasoning about soft constraints and conditional preferences: complexity results and approximation techniques , 2003, IJCAI.

[6]  Andrew Coles,et al.  LPRPG-P: Relaxed Plan Heuristics for Planning with Preferences , 2011, ICAPS.

[7]  Austin Tate,et al.  Semantic Support for Visualisation in Collaborative AI Planning , 2005 .

[8]  James F. Allen,et al.  TRAINS-95: Towards a Mixed-Initiative Planning Assistant , 1996, AIPS.

[9]  Chih-Ping Wei,et al.  Feature Selection for Medical Data Mining: Comparisons of Expert Judgment and Automatic Approaches , 2006, 19th IEEE Symposium on Computer-Based Medical Systems (CBMS'06).

[10]  Andrew Coles,et al.  Temporal Planning with Preferences and Time-Dependent Continuous Costs , 2012, ICAPS.

[11]  Allen P. Nikora,et al.  Automated Identification of LTL Patterns in Natural Language Requirements , 2009, 2009 20th International Symposium on Software Reliability Engineering.

[12]  Julie A. Shah,et al.  Apprenticeship Scheduling: Learning to Schedule from Human Experts , 2016, IJCAI.

[13]  Yu Zhang,et al.  Proactive Decision Support using Automated Planning , 2016, ArXiv.

[14]  James F. Allen,et al.  TRIPS: An Integrated Intelligent Problem-Solving Assistant , 1998, AAAI/IAAI.

[15]  Dawn E. Wilkins Using the SIPE-2 Planning System: A Manual for SIPE-2 , 1999 .

[16]  Jakob Nielsen,et al.  Enhancing the explanatory power of usability heuristics , 1994, CHI '94.

[17]  Jorge A. Baier,et al.  A Heuristic Search Approach to Planning with Temporally Extended Preferences , 2007, IJCAI.

[18]  Alfonso Gerevini,et al.  Preferences and soft constraints in PDDL3 , 2006 .

[19]  Barbara Hayes-Roth,et al.  Human Planning Processes , 1980 .

[20]  Maria Fox,et al.  VAL: automatic plan validation, continuous effects and mixed initiative planning using PDDL , 2004, 16th IEEE International Conference on Tools with Artificial Intelligence.

[21]  Tom Bylander,et al.  Complexity Results for Planning , 1991, IJCAI.

[22]  Joe Marks,et al.  Human-Guided Simple Search , 2000, AAAI/IAAI.

[23]  Subbarao Kambhampati,et al.  AI-MIX: Using Automated Planning to Steer Human Workers Towards Better Crowdsourced Plans , 2014, HCOMP.

[24]  Matthias Scheutz,et al.  What to do and how to do it: Translating natural language directives into temporal and dynamic logic representation for goal management and action execution , 2009, 2009 IEEE International Conference on Robotics and Automation.

[25]  Mary L. Cummings,et al.  Human-automated planner collaboration in complex resource allocation decision support systems , 2010, Intell. Decis. Technol..

[26]  Austin Tate Mixed Initiative Planning in O-Plan2 , 2004 .

[27]  Maria Fox,et al.  PDDL2.1: An Extension to PDDL for Expressing Temporal Planning Domains , 2003, J. Artif. Intell. Res..

[28]  Karen L. Myers Advisable Planning Systems , 1996 .

[29]  Rosemary Luckin,et al.  Modelling Human Teaching Tactics and Strategies for Tutoring Systems: 14 Years On , 2015, International Journal of Artificial Intelligence in Education.

[30]  Hema Raghavan,et al.  Active Learning with Feedback on Features and Instances , 2006, J. Mach. Learn. Res..