Robot planning improved by legal reasoning with uncertainty handling
Legislating World-Model-Based Planning with Legal Reasoning
RoboticsArtificial IntelligenceLogic in Computer Science
Summary
Robots need rules to behave safely and fairly in human spaces, but it’s hard to turn legal texts into clear instructions for robots. The authors built a system that uses legal logic together with a robot’s learned understanding of the world to plan actions while avoiding rule violations from the start. They tested it on a simulated robot arm moving a cube, showing it followed rules more often and handled perception errors better. Their work highlights challenges in precisely linking laws to robot behaviors but shows a path toward safer, law-abiding robots.
What this means in practice
- •For industrial robot developers: Integrate legal constraints into robot motion planning to prevent unsafe or illegal actions before execution.
- •For autonomous vehicle engineers: Incorporate legal reasoning with perception uncertainty to improve decision making aligned with traffic laws.
Authors
Dylan Waldner, Yiannis Kantaros, Guido Governatori, Risto Miikkulainen, Amir Banifatemi
Abstract
As robotic systems grow more general, legal norms are needed to integrate them into society. This paper extends the isomorphism problem of aligning legal source texts with their encodings, and measures two key challenges to robot normative control: (1) the \textit{grounding isomorphism gap}, where perception error grounds false atoms for legal reasoning, and (2) the \textit{ontological isomorphism gap}, where one legal conclusion admits many faithful translations into planning constraints. The paper introduces a legal planning stack that employs Defeasible Deontic Logic (DDL) to constrain a motion planner. The stack leverages learned world models to plan and to provide legal context, enabling \textit{ex ante} governance that intervenes before an illegal action is executed. It was deployed on a simulated robot arm pushing a cube across a $3\times3$ grid. The findings were (1) the legislated agent abided substantially more often than the non-legislated one, and modeling perception uncertainty lifted abidance even further, (2) the legal reasoning ran efficiently at runtime and its verdicts were auditable, and (3) the stack adapted to exogenous signals and endogenous rule changes. Both gaps were measured: (4) world model and probe error corrupted the factual input for the DDL reasoner, and (5) a single law admitted several faithful metric interpretations yielding drastically different abidance. Thus, \textit{ex ante} legislation functions as intended, and closing these gaps with a standardized mapping from the law to runtime constraints and improved fact grounding from perception will yield robust laws that align robot behavior with society's norms.