Dexterous Robotic Grasping in Space: Challenges and Current Research

*Call for Posters - Deadline Extension*

The deadline for poster submission has been extended to 15th November 2025. We are still accepting 1-page posters, showcasing state-of-the-art in space grasping.  Please send your posters to nikolaos.mavrakis@durham.ac.uk and use "iSpaRo 2025 Dexterous Grasping Poster" as your email subject.

*Call for Posters*

We welcome posters from students and researchers that showcase state-of-the-art research in robotic grasping for space. The poster size should be A0, single-page and on a topic relevant to robotic grasping in space, either for orbital or planetary applications. Potential topics include, but not limited to:

• Grasp synthesis 

• Grasping mechanisms

• Teleoperation for grasping

• V&V for space grasping

• Learning and AI for grasping

• Vision and perception for grasping

• Soft and bio-inspired robotic grasping

• Space grasping applications (e.g. satellite servicing, debris capturing, assembly, planetary construction)

Deadline: Please send your posters to nikolaos.mavrakis@durham.ac.uk by October 25th 2025. Please use "iSpaRo 2025 Dexterous Grasping Poster" as your email subject. The selected posters will be notified by November 1st 2025.

About the Workshop

Dexterous grasping remains one of the most challenging areas in robotics research, with significant open problems including in-hand manipulation, stable grasp synthesis, and task-informed grasping. While terrestrial robots are capable of grasping and manipulating a wide variety of objects across diverse settings, space robots still lag behind in dexterity. This disparity is evident in examples such as:

- In orbit, object capture is typically achieved using interfacing docking plates rather than dexterous manipulators. By contrast, terrestrial robots employ grippers and robotic hands to grasp objects.

- On Earth, there is a wide variety of robotic hands, ranging from open-source designs and academic prototypes to fully deployed industrial systems. In space, only a limited number of robotic hands have been prototyped and tested, with the Robonaut hand being a notable example.

- Terrestrial grasping is supported by well-established datasets, protocols, and benchmarking methodologies, whereas in space such standards have only recently begun to emerge.

However, upcoming space markets and applications will demand greater autonomy and dexterity to support operations such as in-space assembly and manufacturing, orbital debris capture, and astronaut assistance. Future planetary missions will require manipulation and grasping capabilities that exceed those currently available. The importance of dexterous manipulation in space has been formally recognized in the most recent NASA Technology Taxonomy (TX04.3.1 and TX04.3.2, Dexterous Robotic Manipulation and Grappling Technologies).

The workshop will bring together experts in space robotics, grasping, and manipulation from academia, industry, and space organizations. Dexterous robotic grasping in space is not as widely adopted as on Earth applications. The goal of the workshop would be to define the challenges of grasping operations in orbital and planetary environments, set clear pathways for the development, testing, and benchmarking of space-based robotic grasping, and describe key user cases in space that can benefit from robotic grasping.

At the same time, the workshop aims to create a community of researchers in the niche field of space grasping and manipulation, with the intention of stimulating interaction between researchers, update the audience on the progress of key players and agencies in the field of space grasping, and showcase state-of-the-art work from young researchers.

Invited Speakers

Program

• 09:00 – 09:15: Welcome and Introduction

• 09:15 – 09:35: Marie Farrell - Formal Modelling and Runtime Verification for Autonomous Grasping in Active Debris Removal 

• 09:35 – 09:55: Maximo Roa - Space Robotic Manipulation Technology at DLR

• 09:55 – 10:15: Carol Martinez - Perception for Intelligent Robotic Grasping in Space

• 10:15 – 10:45: Break and Poster Session

• 10:45 – 11:05: Shaun Azimi - Reliable Dexterous Robotic Systems for Human Space Exploration

• 11:05 – 11:25: Carl “Glen” Henshaw - Show Me The Data: A Rant About Tactile Sensing, Foundation Models, and Neural Network Architectures

• 11:25 – 11:55: Panel Session

• 11:55 – 12:00: Closing Remarks

Invited Talk Abstracts

• Marie Farrell - Formal Modelling and Runtime Verification for Autonomous Grasping in Active Debris Removal 

In this talk, I will provide an overview of using a suite of verification approaches to verify correct behaviour for an autonomous grasping algorithm. This will include requirements elicitation and formalisation using NASA’s Formal Requirements Elicitation Tool (FRET), static verification with the Dafny program verifier and runtime verification using the ROSMonitoring framework. I will discuss the development and verification process, as well as the experiments and simulations that were constructed.  

• Maximo Roa - Space Robotic Manipulation Technology at DLR

DLR has been developing robotic manipulation technologies for over three decades, successfully working both on hardware (robotic manipulators and grippers) and software (torque and compliant control, as well as motion, task and assembly planning). This talk will provide an overview of missions and projects involving robotic manipulation, focusing on aspects related to the design, operation and control of robotic arms and end effectors. 

• Carol Martinez - Perception for Intelligent Robotic Grasping in Space

In this talk, I will present the work of the Space Robotic Research (SpaceR) group in developing robust perception systems for intelligent grasping in space environments. Our work spans classical vision-based methods to emerging tactile and contact-sensing techniques. I will highlight key challenges and discuss ongoing and future research aimed at using these modalities to enable more reliable and adaptive manipulation. Enhancing perception-driven autonomy for space robotics. 

• Shaun Azimi -  Reliable Dexterous Robotic Systems for Human Space Exploration

NASA , together with commercial and international partners, is planning to establish a sustained presence on the lunar surface as part of the Artemis campaign. Dexterous robotics is a key technology area which will enable remote operations on the lunar surface in applications such as logistics, maintenance, science utilization, construction, assembly, and outfitting. The lunar surface will serve as a proving ground for robotics with increasing levels of autonomy, flexibility, and resilience that will enable the human exploration of Mars. I will be giving a brief overview of the challenges associated with dexterous robotic remote operations in architectures designed around human explorers and the plans that NASA is developing to bridge the gap between terrestrial innovation and the demands of space applications. 

• Carl "Glen" Henshaw - Show Me The Data: A Rant About Tactile Sensing, Foundation Models, and Neural Network Architectures 

Foundation models based on large datasets are taking over the world, including the world of robotic manipulation. However, to date there are no extant datasets that include touch data, the primary sense upon which human manipulation depends, because there are few (if any) affordable, reliable touch sensors for robots. Also, foundation models are based on transformers, which are optimized for time series data, not temporospatial data, which is downstream from the lack of good data sets. Without becoming material scientists, how might we address these problems? 

Venue

Room A 

Sendai International Center

Sendai, Japan

Organisers