{"slug": "giving-robots-the-sense-of-touch", "title": "Giving Robots the Sense Of Touch", "summary": "Amazon researchers and the University of Michigan developed HydroShear, a physics-based simulator that models tactile shear forces to train robots for complex manipulation tasks. The simulator achieves a 93 percent average success rate in real-world tasks by accurately tracking force history and slip dynamics, overcoming limitations of existing tactile simulators.", "body_md": "| Giving Robots the Sense Of Touch |\n\n| Written by Lucy Black | |||\n| Friday, 17 July 2026 | |||\n|\nHydroShear is a new physics-based simulator that teaches robots how to use their sense of touch to perform complex manipulation tasks. The research aims to improve the way robots are trained to pick up and manipulate objects. Ideally, a robot needs to be able to feel the forces on their fingertips to manipulate objects effectively. Existing digital simulators were either too slow or too simple, so failed to mimic what happens when a robot tries to pick up an object in terms of the friction, stretching, and slipping. Existing tactile simulators have problems. Physics-based methods like finite-element methods accurately model contact forces but are too slow for training reinforcement learning policies at scale. Faster approximations, on the other hand, oversimplify how forces build up and change during contact. They miss critical events like the moment a gripped object begins to slide or the way a soft sensor deforms over time. Vision-based tactile sensors, in which cameras embedded in soft fingertips capture contact geometry, can be used to show the robot the shape of the object, but it's still difficult to work out the forces needed to manipulate the object. What is needed is a way to model tactile shear, the forces that arise when an object slides or rotates against a sensor. HydroShear, a joint project by Amazon researchers and the University of Michigan, gives simulators the ability to accurately model tactile forces, and lets robots learn using simulations. Once the techniques have been learned, they can transfer to the real world with no modification, achieving a 93 percent average success rate across a range of tasks. HydroShear uses a technique known as path-dependent force tracking. Rather than looking at a single instant of contact, it tracks the continuous history of motion and remembers how a soft robotic finger pad stretches and deforms over time. It also uses realistic slip modeling, meaning it captures complex physical interactions like stiction, sliding, and full 3D sensor-object contact. This allows a robot to detect exactly when an object is about to slip from its grasp. Stiction is a blend of static and friction, and describes the initial force required to overcome static adhesion and cause a stationary object to start moving across another surface. HydroShear also has GelSight integration, acting as a digital twin for vision-based tactile sensors like the GelSight Mini. This enables it to map how camera-based soft-skinned grippers deform under physical pressure. When a robot grasps an object and moves it, different points on the object's surface come into contact with the sensor at different times. HydroShear tracks each of these contact points individually, computing how the soft elastomer deforms as the object moves. It then converts these deformations into realistic force fields, accounting for friction, slipping, and the elastomer's material properties. The simulator handles full 3-D motion, and can also make use of parallel GPUs, enabling efficient large-scale policy training. The researchers calibrate HydroShear by collecting controlled real-world data with a robot arm and vision-based GelSight Mini tactile sensors. The calibration isolates four key parameters: how forces dissipate across the sensor surface, how tangential and normal forces build up, and the friction coefficient between the object and the elastomer. The team evaluated HydroShear on four \"contact-rich manipulation tasks\", each highlighting different challenges. The first challenge is peg insertion, in which the robot has to grasp a cylindrical peg at an unknown orientation and insert it into a tight socket. Because the grasp pose varies with each trial, the robot must use tactile feedback alone to detect and correct alignment errors during insertion. The next task is bin packing, in which the robot inserts a cube into a target slot within a crowded bin. Neighboring cubes partially block the slot, so the robot must push through multiobject contact while sensing forces from multiple directions simultaneously. In the book shelving challenge, the robot inserts a book laterally into a shelf, with gravity pulling perpendicular to the insertion direction. The book is larger than the fingertip, producing broad contact patches that make it difficult to localize the object from touch alone. Finally, the robot has to open a drawer while \"external-force perturbations\" are applied at random times. The robot must detect when the handle begins to slip and tighten its grip just enough to maintain hold without crushing it. The researchers say that accurate tactile simulation provides a better way to train robots - train policies entirely in simulation, then deploy them on real robots - as this is dramatically faster and cheaper than learning from real-world interactions, which can damage sensors and require extensive trial and error. For warehouse automation, the approach is particularly valuable. Tasks like bin packing, sorting, and careful handling require robots to feel their way through complex interactions. HydroShear enables robots to learn such skills without extensive real-world data collection. The research paper is available to download now, and the HydroShear code, simulation demos, and real-world evaluation videos are available on the project page. ## More Information\n## Related Articles\nTo be informed about new articles on I Programmer, sign up for our\n## Comments\nor email your comment to: |\n|||\n| Last Updated ( Friday, 17 July 2026 ) |", "url": "https://wpnews.pro/news/giving-robots-the-sense-of-touch", "canonical_source": "http://www.i-programmer.info/news/169-robotics/19014-giving-robots-the-sense-of-touch.html", "published_at": "2026-07-17 15:44:04+00:00", "updated_at": "2026-07-17 16:06:44.297478+00:00", "lang": "en", "topics": ["robotics", "artificial-intelligence", "machine-learning", "ai-research", "ai-products"], "entities": ["Amazon", "University of Michigan", "HydroShear", "GelSight Mini"], "alternates": {"html": "https://wpnews.pro/news/giving-robots-the-sense-of-touch", "markdown": "https://wpnews.pro/news/giving-robots-the-sense-of-touch.md", "text": "https://wpnews.pro/news/giving-robots-the-sense-of-touch.txt", "jsonld": "https://wpnews.pro/news/giving-robots-the-sense-of-touch.jsonld"}}