Technology Category
Search and Optimization: Sampling/Simulation-based SearchIntelligent Robots: Learning & Optimization for ROBConstraint Satisfaction and Optimization: Constraint Optimization
CACTO-BIC: Scalable Actor-Critic Learning via Biased Sampling and GPU-Accelerated Trajectory Optimization
🤖 AI Summary
This work proposes an efficient continuous Actor-Critic framework to address the high computational cost and poor scalability of integrating trajectory optimization with reinforcement learning in high-dimensional systems. By leveraging properties of the value function under locally optimal policies, the method introduces a sampling strategy biased toward the initial state distribution. Furthermore, it pioneers the integration of GPU-accelerated trajectory optimization into the CACTO framework, substantially improving both data efficiency and computational speed. Experimental results demonstrate that the proposed approach outperforms CACTO and PPO in terms of sample efficiency and convergence rate, and successfully enables real-time high-dimensional control on the AlienGO quadrupedal robot.
Technology Category
Application Category
📝 Abstract
Trajectory Optimization (TO) and Reinforcement Learning (RL) offer complementary strengths for solving optimal control problems. TO efficiently computes locally optimal solutions but can struggle with non-convexity, while RL is more robust to non-convexity at the cost of significantly higher computational demands. CACTO (Continuous Actor-Critic with Trajectory Optimization) was introduced to combine these advantages by learning a warm-start policy that guides the TO solver towards low-cost trajectories. However, scalability remains a key limitation, as increasing system complexity significantly raises the computational cost of TO. This work introduces CACTO-BIC to address these challenges. CACTO-BIC improves data efficiency by biasing initial-state sampling leveraging a property of the value function associated with locally optimal policies; moreover, it reduces computation time by exploiting GPU acceleration. Empirical evaluations show improved sample efficiency and faster computation compared to CACTO. Comparisons with PPO demonstrate that our approach can achieve similar solutions in less time. Finally, experiments on the AlienGO quadruped robot demonstrate that CACTO-BIC can scale to high-dimensional systems and is suitable for real-time applications.
Problem
Research questions and friction points this paper is trying to address.
Scalability
Trajectory Optimization
Reinforcement Learning
Optimal Control
High-dimensional Systems
Innovation
Methods, ideas, or system contributions that make the work stand out.
biased sampling
GPU acceleration
trajectory optimization
actor-critic
scalable reinforcement learning
🔎 Similar Papers
No similar papers found.
