Process
Task planning and execution
You might say: behaviour logic, mission
Task planning and execution sequences skills such as navigation, detection, grasping, and placement, and handles their outcomes.
- What it does
- Decides what the robot should do next and what recovery to try after failure.
- Ask
- What is the recovery path when a skill fails or the world no longer matches the plan?
What to look for
- Navigation, detection, grasping and placement sequence
- Recovery after a skill failure
- Next skill selected for the robot
Catalogue: Executive
- Finite state machine: An explicit set of states and transitions; it is easy to inspect but can become tangled as cases grow.
- Behavior tree: A tree of sequences, fallbacks, and actions is ticked from the root, allowing modular and reactive execution.
- Symbolic planner (PDDL): A symbolic planner searches action models with preconditions and effects for a sequence that reaches a goal state.
Sources
- Nav2, Behavior Trees, Behavior Trees
It creates a more scalable and human-understandable framework for defining multi-step or many state applications.
- Michele Colledanchise and Petter Ögren, Behavior Trees in Robotics and AI, Introduction
A Behavior Tree (BT) is a way to structure the switching between different tasks in an autonomous agent.
- Steven M. LaValle, Planning Algorithms, Ch. 2 §2.4, logic-based planning
For many discrete planning problems that we would like a computer to solve, the state space is enormous.
Use it on your project
teardown-apply: hold "Task planning and execution" (task-planning) from the teardown at https://teardownatlas.com/robotics/ against my project
Paste it into a coding agent that has the teardown skills.
An AI agent wrote this map with the teardown skill. It can contain errors: check the sources before you rely on a part. About and legal