Collaborative Object Transportation in Space via Impact Interactions

Joris Verhagen, Jana Tumova,
KTH Royal Institute of Technology
RSS 2025
Paper Code
Teaser Video

The absense of gravity allows for a new way of thinking about object transportation.
The robots can use impact interactions to control the object.

Abstract

We present a planning and control approach for collaborative transportation of objects in space by a team of robots. Object and robots in microgravity environments are not subject to friction but are instead free floating. This property is key to how we approach the transportation problem: the passive objects are controlled by impact interactions with the controlled robots. In particular, given a high-level Signal Temporal Logic (STL) specification of the transportation task, we synthesize motion plans for the robots to maximize the specification satisfaction in terms of spatial STL robustness. Given that the physical impact interactions are complex and hard to model precisely, we also present an alternative formulation maximizing the permissible uncertainty in a simplified kinematic impact model. We define the full planning and control stack required to solve the object transportation problem; an offline planner, an online replanner, and a low-level model-predictive control scheme for each of the robots. We show the method in a high-fidelity simulator for a variety of scenarios and present experimental validation of 2-robot, 1-object scenarios on a freeflyer platform.

Method

We present the following method for the object transportation problem:

  1. Offline Planner: An offline Mixed-Integer Program generates kinematically feasible trajectories satisfying a high-level STL specification. The planner resolves where, when and with whom objects should impact to fulfill this specification.
  2. Online Replanner: The replanner updates the trajectories based on the current state of the system. It considers the updated positions and velocities of the objects and the physical sizes of objects and robots.
  3. Low-level Controller: The low-level controller tracks the motion plan and controls the robots. It uses an impact-aware model-predictive control scheme to ensure that the robots follow the planned trajectories.
Bezier curves

We parametrize trajectories for objects as 1st-order Bézier curves and robots as dth-order Bézier curves.

Software-In-The-Loop Corridor Scenario

An intuitive example: transport a free-floating object from an initial state to a goal state (green) while avoiding obstacles (red). Two robots are used to transport the object by using impact interactions. We present the STL planner with spatial robustness and impact robustness. Spatial robustness maximizes (informally) the distance to obstacles and the distance inside areas of interest. Impact robustness maximizes the permissible uncertainty in the post-impact velocity of the object, capturing the uncertainty in the impact model.

Spatially robust scenario

Spatially robust corridor travel scenario.

Impact robust scenario

Impact robust corridor travel scenario.

Hardware validation

We perform the corridor scenario on the ATMOS platform at KTH. We additionally show a scenario with a more complex STL specification: the object should visit each region within the time horizon.

Spatially robust scenario

Spatially robust corridor travel scenario.

Impact robust scenario

Spatially robust pong scenario. (a): planned trajectories, (b): executed trajectories.

BibTeX

@article{verhagen2025collaborative,
        title={Collaborative Object Transportation in Space via Impact Interactions},
        author={Verhagen, Joris and Tumova, Jana},
        journal={arXiv preprint arXiv:2504.18667},
        year={2025}
      }