A Conflict-driven Interface between Symbolic Planning and Nonlinear Constraint Solving

TL;DR

This paper introduces a novel iterative algorithm that effectively integrates symbolic planning with nonlinear constraint solving, improving task and motion planning efficiency in robotics.

Contribution

It presents a new bidirectional interface connecting logic planning and nonlinear optimization, along with the PNTC formulation for modeling TAMP problems.

Findings

Framework significantly outperforms alternative approaches

Efficient detection of infeasible constraint subsets

Improves integration of logic and continuous optimization

Abstract

Robotic planning in real-world scenarios typically requires joint optimization of logic and continuous variables. A core challenge to combine the strengths of logic planners and continuous solvers is the design of an efficient interface that informs the logical search about continuous infeasibilities. In this paper we present a novel iterative algorithm that connects logic planning with nonlinear optimization through a bidirectional interface, achieved by the detection of minimal subsets of nonlinear constraints that are infeasible. The algorithm continuously builds a database of graphs that represent (in)feasible subsets of continuous variables and constraints, and encodes this knowledge in the logical description. As a foundation for this algorithm, we introduce Planning with Nonlinear Transition Constraints (PNTC), a novel planning formulation that clarifies the exact assumptions our algorithm requires and can be applied to model Task and Motion Planning (TAMP) efficiently. Our experimental results show that our framework significantly outperforms alternative optimization-based approaches for TAMP.