SOURCE-LINKED INTELLIGENCE

Coding Agents for Generalized Task and Motion Planning Problems

arXiv · AI, language, vision and robotics · article · Sep 24, 2026 · UTC

Task and motion planning (TAMP) problems remain difficult even with full observability and object-centric states because discrete decisions are tightly coupled to geometric, kinematic, and dynamic constraints. Generalized TAMP addresses this difficulty by exploiting regularities across problem instances to reduce planning effort on new instances. However, existing methods require substantial TAMP-specific engineering. We investigate whether coding agents can automate this process by synthesizing programs that generalize across instances. Given a task description and simulator access, each agen

Read original source ↗ Open in workspace

recordType

paper

region

Global