Perception for Action

Following this line of research we investigated the effects of action on perception and vice versa. Our main focus was on perception during locomotion. One question for example is: "how do we perceive our own speed while we are walking?". Most of this work is conducted in the frame of the European Projects "CyberWalk". CyberWalk is a collaborative research project between the Technical University in Munich, the ETH in Zurich, the University of Rome and us. The goal of the CyberWalk project is to enable unconstrained walking through virtual worlds. At its heart CyberWalk is developing an omni-directional treadmill which allows us to walk in any direction through virtual worlds without actually moving through space.

Walking in Virtual Worlds

Despite recent improvements in Virtual Reality technology hardly any satisfactory solutions exist that enable users to physically walk through virtual environments in a natural way. In the CyberWalk project our goal is to enable unconstrained, omni-directional walking in virtual worlds. To achieve this goal we follow a holistic approach that unites science, technology and applications. CyberWalk will develop a novel omni-directional treadmill. The concept of motion control behind the treadmill will focus on diminishing the forces exerted on the walking user, by minimizing the overall accelerations. The CyberWalk project will showcase its developments via a physical walkthrough through the virtual reconstruction of the ancient city of Pompeii. However, there are many other application areas for which the CyberWalk approach might proove useful, e.g., medical treatment and rehabilitation (Parkinson’s disease, phobia, etc.), entertainment, sports (training facilities, fitness centers), behavioural science, education, training maintenance teams, security guards, etc.), and architecture (exploring large virtual construction sites).

Perception during Locomotion

The main focus of this research direction is concerning sensory integration during walking. Here we investigate several topics:

Perceived walking speed
How does walking through space differ from walking in place, such as on a treadmill? Signals involved range from visual, vestibular and proprioceptive cues to motor comands sent to the legs for locomotion. For example, unsing a treadmill allows us to study the contribution of vestibular information to perceived walking speed.

Perceived visual speed during walking
Visual motion may be perceived differently during walking and standing (Durgin et al., 2005). The difference in percept has been interpreted as an adaptive change in neural coding. To date, however, the menchanisms of this adaptive change are still unclear. Therefore, we here investigate visual motion perception under different active and passive viewing conditions.

Walking in Circles

According to common belief, people tend to walk in circles when they get lost in for instance the desert or the jungle. This has been attributed to differences in leg length, causing people to walk in the direction of their shorter leg (Lund, 1930). Although quite some research has been done on the ability of people to walk blindfolded to a previously seen location (see e.g. Loomis et al., 1992), the most extensive study about walking for longer distances dates from 1928 and still didn’t include walking with vision in an unknown environment. In collaboration with the national German television channel WDR - Redaktion Kopfball, we decided to investigate whether people really walk in circles and what might be the cause(s) of this. What makes the question of special interest to our research is that here all factors that influence our walking behaviour come together: biomechanics (leg length, leg strength, other body asymmetries), perception (vision, vestibular system, subjective straight-ahead) and cognition (landmark use, cognitive strategies to find ones way, knowledge about environment, decision making).

Statistics of natural Walking Behaviour

How do we walk? Can we describe human walking behaviour in natural environments using statistical measures? Most research on walking is done under controlled laboratory conditions. Recent technological advances, however, make it possible to study walking under natural circumstances as well. To measure natural human walking behaviour we use a combined Global Positioning System (GPS) / Inertial Measurement Unit (IMU) setup. We have collected a large amount of data of persons walking through a suburban environment and now use these data as a benchmark for the evaluation of human locomotion behaviour in real and virtual environments. Most of this work was conducted as a Master Thesis by Manish Sreenivasa in the framework of the CyberWalk project.

Go to Editor View