Smart Suction Cup | Jungpyo Lee

The Smart Suction Cup

The Smart Suction Cup is a multi-chambered end effector in which each internal chamber is connected to a pressure sensor, enabling real-time monitoring of airflow changes during suction actuation. Rather than instrumenting the contact surface with additional tactile sensors, the system treats suction itself as a sensing modality: chamber pressures reveal how well the cup is sealed against an object surface.

This design targets the central challenge in suction grasping—air sealing—without adding electronics or extra sensors on the cup face. The cup body is fabricated as a multi-chamber pneumatic structure; each chamber is plumbed to a remote pressure sensor so that no electronics sit on the contact face. This keeps the end effector easy to manufacture, replace, and use in contact-rich tasks without risking damage to fragile on-cup instrumentation.

The baseline design uses four internal chambers. We have since developed 3- and 6-chamber variants to study how chamber geometry affects sensing resolution and control performance.

Smart Suction Cup fabrication and system integration

Haptic Search

Traditionally, suction grasping has relied heavily on vision to identify optimal grasp locations. Vision-based methods can fail when camera resolution is limited, visual artifacts are present, or object surfaces are complex or adversarial.

When a vision-based grasp planner proposes a pose that fails to form a reliable seal, internal pressure measurements indicate where air leakage occurs. The robot can then adjust its pose toward regions of higher vacuum, using tactile (pressure) feedback to search for a better grasp point. We call this process the haptic search algorithm.

By measuring airflow in each chamber, haptic search complements vision in settings where visual grasp estimates are unreliable—recovering from failed seals and guiding the robot toward poses that improve suction performance.

As shown in the videos below, when the cup is near an object edge, haptic search drives lateral motion to move inward toward the object surface. When rotational alignment is off, the robot adjusts its orientation to match the surface normal. These adjustments refine the robot pose until suction sealing improves.

Demonstrations

Bin-picking with adversarial objects

We implemented haptic search in a pick-and-place bin-picking task using 19 adversarial objects deliberately chosen to be hard to capture with a vision system. Guided by internal flow measurements, model-based haptic search improved grasp success by up to 2.5× compared with a state-of-the-art vision-only baseline (Dex-Net 4.0). This work was published in IEEE Transactions on Robotics (T-RO) and was selected as an Editor’s Choice article by Science Robotics.

PCB regrasping on a conveyor belt system

Building on the same haptic search framework, we applied the Smart Suction Cup to regrasping printed circuit boards (PCBs) in an industrial recycling context—a collaborative project with Prof. Fei Chen at CUHK (as required by innoHK's collaboration policy). Waste PCBs are difficult to grip reliably because of uneven surfaces and unmodeled surface-mounted components that disrupt suction sealing. We studied regrasping on both stationary setups and conveyor belt systems with realistic vision-based grasp planners, demonstrating improved pick-and-place success in an industrial recycling workflow. This work was published at ICRA 2024.

Object contour following

Beyond recovering from failed grasps, we extended the system to contour following—using chamber pressure feedback to track edges and planar contours on object surfaces. Because the sensor relies on internal airflow rather than vision or wear-prone contact transducers, it remains useful under occlusions and during sustained sliding contact. Experimental validation showed robust edge tracking at speeds up to 3 cm/s. This work was published at IROS 2024.

Co-design of suction cup structure and haptic search control

We are extending the Smart Suction Cup along two coupled directions: multi-chamber mechanical design and haptic search control policy. Different chamber layouts change what leakage patterns the sensors can observe, which in turn changes what control strategies are most effective.

To support this, we built a simulation environment for multi-chambered suction cups and collected data to study how sensor geometry and control policy should be designed together. The goal is co-design: jointly optimizing cup structure and haptic search controllers to improve adaptability across grasping tasks. This line of work is currently under review.

Under Review

Co-designing suction cup structure and haptic control policy for adaptive robotic grasping

Jungpyo Lee and Hannah S Stuart
TRO

Haptic search with the Smart Suction Cup on adversarial objects

Jungpyo Lee^*, Sebastian D Lee^*, Tae Myung Huh, and 1 more author

IEEE Transactions on Robotics, 2024

Spotlighted article in Science Robotics Editor’s Choice.

Abs DOI PDF Video

Suction cups are an important gripper type in industrial robot applications, and the prior literature focuses on using vision-based planners to improve grasping success in these tasks. Vision-based planners can fail due to adversarial objects or lose generalizability for unseen scenarios, without retraining learned algorithms. In this article, we propose haptic exploration to improve suction cup grasping when visual grasp planners fail. We present the smart suction cup, an end effector that utilizes internal flow measurements for tactile sensing. We show that model-based haptic search methods, guided by these flow measurements, improve grasping success by up to 2.5× as compared with using only a vision planner during a bin-picking task. In characterizing the smart suction cup on both geometric edges and curves, we find that flow rate can accurately predict the ideal motion direction even with large postural errors. The smart suction cup includes no electronics on the cup itself, such that the design is easy to fabricate and haptic exploration does not damage the sensor. This work motivates the use of suction cups with autonomous haptic search capabilities in especially adversarial scenarios.
Accepted

Enhanced Haptic Search with a Multi-chamber Suction Cup using Reverse Flow

Gia Jeong, Jungpyo Lee^†, and Hannah S Stuart
IROS

Haptic Contour Following with the Smart Suction Cup

Sebastian Lee, Jungpyo Lee, and Hannah S Stuart

In 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024

Abs DOI PDF Video

The Smart Suction Cup is a tactile sensing and gripping system designed to enhance pick-and-place operations in industrial settings. While previous research has primarily focused on utilizing this technology for haptic search in cases of initial grasp failure, this study introduces a novel application: following contours. This function is already established as an important function for object recognition and grasp planning – substantiated by numerous works using other tactile sensors. Here, we explore contour following for a flow-based tactile sensor because it is not susceptible to visual occlusions nor tactile sensor wear. Experimental validation demonstrates the Smart Suction Cup’s ability to track edges at different speeds and navigate various planar contours, showcasing rapid and robust tracking of edges. Notably, the Smart Suction Cup can reliably operate at a speed of 3 cm/s. This is one step towards the adoption of the Smart Suction Cup for real-world applications.
ICRA

Regrasping on Printed Circuit Boards with the Smart Suction Cup

Jungpyo Lee, Zheng Sun, Zhipeng Dong, and 2 more authors

In 2024 IEEE International Conference on Robotics and Automation (ICRA), 2024

Abs DOI PDF Video

The disposal of waste electrical and electronic equipment (WEEE) presents a sustainability challenge, particularly for waste printed circuit boards (PCBs). PCBs are challenging to sort out from other waste materials in part because traditional industrial end-effectors struggle to reliably grip these irregularly shaped objects with unmodeled surface-mounted components. Vision-based separators, while effective for object categorization, face challenges with identifying precise grasp points on PCB surfaces. This paper studies regrasping control to enhance suction cup grasping performance on PCBs, addressing issues arising from uneven surfaces and intricate features that interfere with suction sealing. We categorize PCBs into two recycling levels – with large surface features intact or removed – and conduct experiments on both stationary and conveyor belt setups with realistic vision-based grasp planners. Results show that jumping regrasping improves pick-and-place success rate. Haptically driven jumping – using the Smart Suction Cup – is especially useful for unprocessed waste PCBs with large surface mount parts. The proposed method offers a promising solution to enhance the efficiency and reliability of robotic grasping in recycling applications.