Manipulate-to-Navigate: Reinforcement Learning with Visual Affordances and Manipulability Priors
🤖 AI Summary
To address navigation failure caused by movable obstacles blocking paths in dynamic environments, this paper proposes a “Manipulate-to-Navigate” framework enabling end-to-end co-optimization of manipulation and navigation. Methodologically, it integrates operability priors with vision-based utility maps to guide policy learning, significantly improving sample efficiency and action-value estimation of reinforcement learning in complex scenes; policies are trained in simulation and deployed on a Boston Dynamics Spot robot, demonstrating cross-task generalization. Contributions include: (1) the first reinforcement learning paradigm jointly optimizing manipulation and navigation for mobile manipulators; (2) an operability-aware visual utility representation mechanism; and (3) empirical validation on newly introduced Reach and Door benchmark tasks, with successful sim-to-real transfer of the Reach policy to the physical Spot platform, achieving robust interactive navigation in dynamic environments.
Application Category
📝 Abstract
Mobile manipulation in dynamic environments is challenging due to movable obstacles blocking the robot's path. Traditional methods, which treat navigation and manipulation as separate tasks, often fail in such 'manipulate-to-navigate' scenarios, as obstacles must be removed before navigation. In these cases, active interaction with the environment is required to clear obstacles while ensuring sufficient space for movement. To address the manipulate-to-navigate problem, we propose a reinforcement learning-based approach for learning manipulation actions that facilitate subsequent navigation. Our method combines manipulability priors to focus the robot on high manipulability body positions with affordance maps for selecting high-quality manipulation actions. By focusing on feasible and meaningful actions, our approach reduces unnecessary exploration and allows the robot to learn manipulation strategies more effectively. We present two new manipulate-to-navigate simulation tasks called Reach and Door with the Boston Dynamics Spot robot. The first task tests whether the robot can select a good hand position in the target area such that the robot base can move effectively forward while keeping the end effector position fixed. The second task requires the robot to move a door aside in order to clear the navigation path. Both of these tasks need first manipulation and then navigating the base forward. Results show that our method allows a robot to effectively interact with and traverse dynamic environments. Finally, we transfer the learned policy to a real Boston Dynamics Spot robot, which successfully performs the Reach task.
Problem
Research questions and friction points this paper is trying to address.
Mobile manipulation in dynamic environments with obstacles
Integration of navigation and manipulation for obstacle clearance
Learning feasible manipulation actions to enable subsequent navigation
Innovation
Methods, ideas, or system contributions that make the work stand out.
Reinforcement learning for manipulate-to-navigate tasks
Combines manipulability priors and affordance maps
Reduces exploration with feasible, meaningful actions
🔎 Similar Papers
No similar papers found.
