HATP: Hierarchical Agent-based Task Planner

Abstract

Hierarchical Task Network (HTN) planning is a proven approach to solving complex, real world planning problems more efficiently than planning from first principles when 'standard operating procedures' (or `recipes') can be supplied by the user. By planning for tasks in the same order that they are later executed, total-order HTN planners always know the complete state of the world at each planning step. This enables writing more expressive planning domains than what is possible in partial-order HTN planning, such as preconditions with calls to external procedures. Such features have facilitated the use of total-order HTN planners in agent systems and seen them excel in AI games. This paper describes the Hierarchical Agent-based Task Planner (HATP), a total-order HTN planner. Since its first implementation, HATP has had various extensions and integrations over the years, such as support for splitting a solution into multiple streams and assigning them to the agents in the domain; modelling their beliefs as distinct world states; allowing 'social rules' to be included by the user to define what kind of agent behaviour is appropriate; allowing tasks to be planned by taking the human's safety and comfort into account; and to interleave HTN and geometric planning. Since many of these implementations have remained prototypes, we have significantly enhanced them as well as HATP itself, and integrated them into a stand-alone distribution, which is now available as open source software (under a BSD 2-Clause License). This paper presents some of our recent improvements to HATP, and gives an overview of its user-friendly language, which treats agents as distinct entities; its mechanisms for effective control over decomposition; and its integration into our larger robotics framework.