Research Analyzer
← Back ICRA 2026

Sample-Based Hybrid Mode Control: Asymptotically Optimal Switching of Algorithmic and Non-Differentiable Control Modes

Yilang Liu, Haoxiang You, Ian Abraham

PDF

AI summary

Key figure (auto-extracted from paper)
A sample-based optimization method enables provably convergent, optimal switching between discrete and non-differentiable control modes for agile robotic tasks.
Hybrid control sample-based optimization mode switching agile robotics non-differentiable control robotic locomotion

Problem

Agile robots must dynamically switch between discrete, non-differentiable, and algorithmic control modes, but traditional hybrid control methods struggle with the combinatorial complexity and lack of gradients, leading to instability and suboptimal performance.

Approach

The authors reformulate hybrid mode switching as a discrete-time integer optimization problem and solve it iteratively by uniformly sampling mode transition tuples (mode, start time, duration) to efficiently search the solution space without gradients or predefined schedules.

Key results

  • Novel iterative sample-based formulation for hybrid mode sequencing
  • Provable asymptotic convergence guarantees for mode optimization
  • Successful synthesis of complex agile behaviors on a real-world quadruped
  • Outperforms modern trajectory optimization in synchronizing complex hybrid modes

Why it matters

Enables reliable, high-performance control for agile robots in contact-rich environments, bridging theoretical hybrid control with practical robotic deployment.

Abstract

This paper investigates a sample-based solution to the hybrid mode control problem across non-differentiable and algorithmic hybrid modes. Our approach reasons about a set of hybrid control modes as an integer-based optimization problem where we select what mode to apply, when to switch to another mode, and the duration for which we are in a given control mode. A sample-based variation is derived to efficiently search the integer domain for optimal solutions. We find our formulation yields strong performance guarantees that can be applied to a number of robotics-related tasks. In addition, our approach is able to synthesize complex algorithms and policies to compound behaviors and achieve challenging tasks. Last, we demonstrate the effectiveness of our approach in a real-world robotic examples that requires reactive switching between long- term planning and high-frequency control. Videos are avail- able on https://yilangliu.github.io/hybrid_mode_ sampling/

Index terms

Hybrid Logical/Dynamical Planning and Verification Optimization and Optimal Control Legged Robots

Related papers