When you touch objects in your surroundings, you can discern each item's physical properties from the rich array of haptic cues you experience, including both the tactile sensations arising in your skin and the kinesthetic cues originating in your muscles and joints. For example, reaching out to grasp a water glass refines your visual estimates of its location, size, shape, and weight while also making you rapidly aware of the temperature, stiffness, smoothness, and friction of its surfaces. Feeling how these haptic sensations develop in response to your motions enables you to not only perceive the glass's material properties but also manipulate it fluidly, whether your goal is to bring it to your lips to drink, place it upside-down in your dishwasher, or rotate it under a flow of hot water as your other hand scrubs it clean.
Over the course of life, humans leverage their rich sense of touch to master a wide variety of physical tasks, from everyday necessities like buttoning a jacket to difficult feats such as sculpting a marble statue or inserting a needle into a patient's vein. Many tasks are challenging when first tried, but practice usually enables one to improve the resulting interaction and optimize one's motions so they become almost automatic. You can gain some appreciation for the complexity of tasks that normally feel effortless, such as slicing bread or brushing your teeth, by trying to complete them with your non-dominant hand. Similarly, even the simplest manual skills become almost impossible if you lose your tactile sensitivity due to local anesthetic or a lack of blood flow. The crucial role of the sense of touch is also deeply appreciated by researchers working to create autonomous robots that can competently manipulate everyday objects and safely interact with humans in unstructured environments. Such systems rarely take advantage of haptic cues and thus often struggle to match the perception, manipulation, and interaction capabilities of humans.
Although humans experience touch coherently, this sense stems from a wide range of distributed receptors that each responds most strongly to a different type of stimulation, generally broken into the categories of mechanical, thermal, and pain sensations. Much about the sense of touch is understood, and many other aspects still need to be investigated. The most notable differences between touch and the more well-understood senses of vision and hearing are that exploring the world through touch requires action and that what one feels greatly depends on how one moves. It is also helpful to consider that vision has high spatial acuity and only moderate temporal acuity, while hearing is the opposite; different aspects of haptic perception lie along this spatiotemporal continuum between vision and hearing. Because we don't yet fully understand haptic interaction, few computer and machine interfaces provide the human operator with high-fidelity touch feedback or carefully analyze the physical signals generated during an interaction, limiting their usability.
The Haptic Intelligence Department of the Max Planck Institute for Intelligent Systems aims to elevate and formalize our understanding of haptic interaction while simultaneously inventing helpful human-computer, human-machine, and human-robot systems that take advantage of the unique capabilities of the sense of touch. We pursue this goal by undertaking research projects in the following four main research fields, each of which is more thoroughly described in its own section below:
- Understanding Tactile Contact: Haptic perception is tightly coupled to movement (action) via the physics of contact. Despite the complexity of these phenomena, people intuitively learn how best to move to extract desired information while accomplishing the task at hand. We seek to disentangle these elements by instrumenting physical interactions carried out by both natural agents (humans) and artificial agents (robots), analyzing the resulting signals with physics-based models and machine learning.
- Haptic Interface Technology: Haptic interfaces are mechatronic systems that modulate the physical interaction between a human and his or her tangible surroundings so that the human can act on and feel a virtual and/or remote environment. How can such systems vividly reproduce the perceptual experience of touching real objects and provide feedback that helps the user improve his or her motor skills? We seek to answer these questions by carefully studying existing technologies and inventing new haptic interfaces.
- Teleoperation Interfaces: Commonly used in minimally invasive robotic surgery and hazardous material handling, telerobotic systems empower humans to manipulate items by remotely controlling a robot. How can such systems support the operator to perform tasks with outcomes that are as good as (or even better than) those accomplished via direct manipulation? We work to create new ways to capture operator input, deliver haptic feedback, and otherwise augment the operator's abilities, and we systematically study how these technologies affect the operator.
- Physical Human-Robot Interaction: To help humans in unstructured everyday environments like hospitals and homes, robots need better haptic interaction skills as well as increased social intelligence. We are working to discover whether and how physical human-robot interaction can benefit humanity by advancing robotic tactile perception and designing, building, and evaluating new physically interactive robots targeted at particular user populations.
The specific projects we pursue take shape through a bottom-up process that draws on the experience and interests of the primary researcher, the expertise and creativity of our director and other members of the department, ideas and expertise from our collaborators both within and outside of MPI-IS, and recent discoveries in haptics and related areas. Most of the resulting projects fit well within one of the above fields; a small number of projects bridge topics or are more remote from these core competencies. Each project typically has a lead researcher who is a research scientist, postdoctoral fellow, doctoral student, or masters thesis student. This person will generally be the first author of resulting publications, and Dr. Kuchenbecker will generally be the last author. Depending on the demands of the project, this pair may be supported by one or more other scientists, visiting professors, research engineers, technicians, student research assistants, summer interns, and/or short-term high-school interns.
Members of the HI department enjoy working in our diverse international research environment. Our current team is gender balanced and hails from thirteen countries around the world. Because we pursue highly interdisciplinary projects, we welcome applications from people in a wide range of fields. Many of our current department members have a background in mechanical engineering, biomedical engineering, electrical engineering, computer science, and/or cognitive science. Dr. Kuchenbecker books a standard one-hour meeting time for each lead researcher and research engineer every week. This time is spent discussing the recent progress, challenges, and future goals for each of the researcher's projects. When she travels, she assigns pairwise meetings between researchers in lieu of these individual research meetings.
We hold group meeting for 90 minutes almost every week in Stuttgart, using video conferencing to enable remote members to participate from Tübingen, Switzerland, the USA, Canada, and many other locations. Our recurring agenda covers personnel changes, publication activities, awards and media attention, third-party funding, internal logistics, recent scientific talks, and upcoming events. Each attendee then spends one to two minutes giving an individual update on his or her recent activities, including both successes and challenges. Group members ask each other questions and share suggestions on how to solve the issues others are facing. As a rotating task, a lab member writes down funny things that are said during each group meeting, and our director emails out an edited version of these quotes along with her typed minutes from the meeting.
Our primary publishing targets are full-length journal articles in engineering- and medicine-focused research fields and top-tier conference papers in fields related to computer science. We use hands-on demonstrations of our technology and peer-reviewed short papers (often works in progress, late-breaking reports, or workshop papers) as stepping stones to longer-format papers, beneficially gathering feedback from other researchers early in the process to increase the chances of high-impact contributions. To achieve good visibility for our activities and give our younger scientists experience formally presenting their research, we also frequently publish papers at good conferences in haptics and robotics.
To help people prepare for conference presentations, thesis defenses, and job talks, the HI Department holds presentation club most weeks. The presenter practices his or her talk, answers a large volume of questions similar to what they might expect in the target presentation venue, and then receives kind suggestions from the group on how the presentation could be made more effective. In other weeks, presentation club includes reports from people who recently attended a conference, an interactive discussion of a particular paper, a brainstorming session for one of our research projects, tips from lab members on tools helpful for research, or a lesson by Dr. Kuchenbecker on effective writing. Presentation club is coordinated by a lab member who serves a six-month term in this position. Other departmental leadership roles that rotate on a six-month basis include Demo Coordinator, Human Subjects Coordinator, Internship Wizard, Poster Coordinator, and Tool Master. Our long-term leadership roles are Director, Department Assistant, IT Wizard, Mechanical Design and Manufacturing Expert, Additive Manufacturing Coordinator, Purchasing Wizard, and Safety Representative. Together, we hope to greatly advance human understanding of touch cues while simultaneously discovering new opportunities for their use in interactions between humans, computers, and machines.