Abstract
The importance of realistic artificial sensation on movement has grown rapidly in this technological society. This paper presents a systematic literature review of locomotion based technologies. The objective is to develop a literature database related to locomotion interface Out of more than 30 papers, only 15 papers fit the selection criteria. Identified literatures were read, analyzed and grouped into three main categories of locomotion based technologies, which includes: (i) medical; (ii) entertainment; and (iii) mechanical. The review concludes and discusses ways that future research can benefit from it.
Keywords: locomotion interface; medical; entertainment; mechanical; virtual environment
Introduction
The importance of realistic artificial sensation on movement has grown rapidly. Conversely, there are no or little efforts to review relevant literatures in locomotion interface. The objective of this paper is to review, analyze and classify existing literature in the works that used locomotion based system. This literature review also investigates research trends and identifies related method, framework and sample of research in this area. More specifically, the following are the three research questions:
(1) What is locomotion interface?
(2) How does locomotion interface function?
(3) What are the field that uses locomotion interface and how it is used?
Locomotion interface is a type of haptic device that allows the user to feel an artificial sensation of physical walking and navigate in a virtual environment. Furthermore, locomotion interface is equipped with three main functions that is to create sense of walking while preserving the user’s true position, enables the user to alter bearing direction and to simulate uneven walking surfaces. Locomotion interface is also widely used in virtual reality gaming as it provides the user a very realistic sensation that ultimately enhance the gaming experience. There are countless fields that uses locomotion interface such as in the medical field, entertainment field, mechanical field and so on. These fields use various simulation devices to conduct their experiments.
This paper is organised as follows. In Section 1, we will describe the use of locomotion in medical field. In Section 2, we will discuss more about locomotion in the entertainment field. Next, we will discuss the use of locomotion in mechanical field in Section 3. Lastly, in Section 4, we will conclude the paper and present future work.
1. Locomotion interface in medical applications
Locomotion Interface for Gait Simulation (Stroke Rehabilitation)
Locomotion interface based technology has been widely used for stroke patient rehabilitation as it improves the patient’s motor control tremendously for the past years (NINDS, 2014). The users do the experiential simulation based on two planar two degree of freedom belt-driven mechanisms which fully decouple the horizontal and vertical motion of two auxiliary footplates which have force sensor mounted underneath of each footplate in order to interchange the end effectors according to the user’s intentions. Gait simulation requires the user to interact with the system using their feet. The model created by Delp et al. has been used to construct muscle-driven gait simulation and additional lower extremity motions, but the musculotendon parameters (e.g., optimal fiber length and pennation angle) are based on the studies of cadavers. The reimagining of the model helps to design the free motion or restraining the user’s displacement towards certain direction. There are two periods which are the stance phase and swing phase. The stance phase starts when the user’s heel strike causing force impact to reach 150% of the user’s weight (Dinh-Son et al., 2015). Hence during that phase, the virtual floor needs to support the user’s weight. Meanwhile, the swing phase indicates the movement of the foot above the floor. To allow the user to feel unrestrained movement, the swinging of the foot should be minimized hence reducing the inertia of the end-effector.
Locomotion interface is divided into treadmills and mechanisms based on footplates. Treadmills are used as they can easily stop the user’s movement as they walk on the belt and thus creating the illusion of large motion ranges. There are also several treadmill-based locomotion interfaces that enables the users to move in arbitrary direction (e.g., Omni-directional Treadmill). Alternatively, system based on footplates allows users to move in vertical direction which helps simulate more complex environment. Impedance controller is used to move footplates with pantograph mechanisms in Gait Master, then it uses a magnetic position sensor to measure the point of the markers located on the user’s feet so that the end-effectors stay underneath them.
The experimental environment presents the mechanical design and components of the locomotion interface. Furthermore, it finds solutions to overcome the obstacle of supporting body weight in the vertical direction. The mechanisms are formulated and highlighted to use high-power haptic control as it has been a real challenge because the user is physically connected to powerful motors and unstable condition may end up in physical injuries. Direct force lag control helps lower the inertia of the platform and enables noise cancelling from the force sensor and the possible tremor of the user which tends to experience stability issue in the system. Lastly, platform rendering is evaluated using pilot experiments. The initial experimentation emphasizes the system time response felt by the user during the swing phase. The stance phase requirements are proven by conducting quality assessment of the virtual floor that supports the user’s weight.
In a nutshell, the system dwells mainly in the parallel mechanisms stimulated using belts that decouple the various degrees of freedom and in the force sensors that deduce the intention of the user. Moreover, parallel mechanisms have proven to have better workload and acceleration capabilities paralleled to their on-going counterparts and decoupled removes the singularity drawback that happens in most parallel systems. Swing phase is still limited as it has to face the static friction and the inertia of the platform. Future development includes the development of passive interface enables the free motion of the user and cancellation algorithm brings the users back to their original position.
2. Locomotion interface in entertainment applications
VirtualSpace – Overloading Physical Space with Multiple Virtual Reality Users
Virtual reality (VR) attempts a lower level of immersion if users can handle by walking around in the physical world. This has been assigned to do walking in the real world. Compared to imitated walking using treadmills and to locomotion techniques such as teleportation or walking in place, real walking leads to higher levels of immersion. By referring to the virtual reality headsets available to users that are able of tracking real walking in a room-sized tracking volume, one of them would expect the real walking to become the main approach to VR. However, looking at experiences available for room-scale VR (e.g. Steam) 1, only a slightly percentage appear to be using real walking. The other one applies locomotion techniques, such as on-the-spot teleportation, even with the decreased experience.
The journalists were argued that it is the physical space requirements of real walking that make it inapplicable for users. For example, the rent of 4×4 m (meter) of the space surpasses the cost of a VR headset and tracking system in a complete three months. To decrease the space requirements of real walking, researchers have intended several techniques. Redirected walking loops long walking paths into a finite tracking volume but it requires very large installations (4x10m). For room-scale installations, different approaches such as test how to reshape virtual experiences to fit a random room shapes, which makes it easier to prepare a real walking experience into an existing environment. 2 This approach is constrained to applications, that can fit to room setups and do not have specific space requirements.
VirtualSpace is a novel system that grant the multiple real walking VR users to share the same way physical space without being awake of each other. VirtualSpace is designed to provide each of the user with an illusion of being in custody of the entire physical space. For example, two users occupy the same 4x4m of the physical space, while being tracked using a VR tracking system (Vive). Both users are in their own way, apart in the virtual environments. (a) The green user is absorbed in a badminton app, meanwhile (b) the blue user plays a Pac-Man game. The key point is that both apps are generalized to the entire physical space, VirtualSpace allows each app to be drafted under the expectation that the user has physical access to the whole 4x4m space. And that is true, as VirtualSpace limits each app to a different non-overlapping tile of space. Client apps manage this in a clear way to their user. The badminton app, for example, makes the user’s virtual opponent rebound the ball always to locations inside the tile directly assigned to this app.
To grant users to finish their narrative and to avoid users from noticing that the system is restraining them, VirtualSpace applies what we call maneuvers. Every maneuver comes with a certain start-up delay that grants all apps to get their users ready and every maneuver takes place at a certain movement speed. In this example, the apps may agree on a three-second delay and a one second transition speed. Clockwise and counterclockwise revolutions are enough in that it never takes more than two revolutions to deliver a user to any tile. Revolutions take time in which is not always adaptable with game action sequences that require fast, large, and unusual movements of a user. To enable such movement sequences, VirtualSpace attempts the focus maneuver and this maneuver grants a single app to temporarily take over most of the tracking space. For example, when the badminton app uses a Focus maneuver to grant its user to hit the birdie in the center of the court in a fast succession of ball exchanges.
VirtualSpace’s main improvement is a technique that grants multiple real walking virtual reality users to be overwhelmed into the same physical space. VirtualSpace accomplishes this by assigning each app to a smaller tile, where the overall tracking space is separated into computationally determined individual tiles. General “maneuvers” allow apps to encourage users to walk across the entire physical space, thereby allowing each app to advance its narrative and to prevent users from noticing that they are restrained to a tile. This strategy enables VirtualSpace to accomplish packings of high density, such as 4 users in 16m2. The main constraint is that even though apps can move users anywhere within the tracking space, by getting there may be subject to a delay. Along the same lines, apps must be able to generally obey when another app asks for a maneuver.
The displayed apps were VR typical isolating experiences. VirtualSpace can naturally be continued to multi-user applications, mobile scenarios, and spaces of different shape or size. Since mainly rotation maneuvers have been carried out, a more even analysis of space still seems possible. Increasing the size of the tracking space while maintaining user density could also improve ratings, as the consequence of real walking might demonstrate higher in larger areas on enjoyment and anticipated confinement.
3. Locomotion Interface in Mechanical Application
Mechanical engineering has a crucial role in manufactured technologies, from making vehicles to household machines. A huge number of locomotion interfaces are derived from exercise machines, some of which are stair steppers, stationary bicycle and treadmills. Others have been precisely created towards a specific locomotion interface approach. Treadmills are normally used in locomotion interfaces to allow users to walk through virtual surroundings, and are broadly used in physical exercise and gait rehabilitation (Hejrati, Crandall, Hollerbach, & Abbott, 2015). The major obstacle for treadmill-style locomotion interfaces is simulating overground walking which involve user’s stability. In order to present realistic vestibular stimulation during acceleration, treadport immersive locomotion interface uses an actuated tether to allow self-selected walking speed. It regulates the user’s location that combined user’s volition, as a belt controller is applied which enables the user to naturally self-select their walking speed depends on the walking speed overground (Hejrati, et al., 2015). This provides a more natural locomotion experience as the belt speed can be manually set the users.
Locomotion interface is broadly use in mechanical field and it has frequently been proposed that the greatest locomotion mechanism for virtual reality is walking. The sense of orientation or distance while walking is better compared to riding in a vehicle. A real walking simulation on slopes, stairs and level ground prove that the proposed locomotion interface lets an average individual to walk naturally on numerous virtual terrains in safety, without feeling disturbances. This interface has several applications, such as virtual reality (VR) navigation, gait analysis and gait rehabilitation. Nevertheless, almost all virtual reality applications are not delivering the proprioceptive response of walking as real-walking locomotion interfaces are thought to such as ease user navigation, produce a better sense of presence than other locomotion interfaces and more natural (Peck, Fuchs, & Whitton, 2012). The CirculaFloor locomotion interface offers sense of walking and therefore possibly develop the user knowledge in training simulators and entertainment (Iwata, Yano, Fukushima, & Noma, 2005). The movable tiles applied a holonomic mechanism to accomplish omnidirectional gesture. This allow users to maintain their position while walking in a virtual world. Since CirculaFloor exploits footpad and treadmill together, it uses a couple of movable tiles to create infinite omnidirectional surface. The combination of a projection display and the CirculaFloor has manage to deliver the decisive sense of presence. This is a great accomplishment in mechanical field as the integrated system could significantly contribute to virtual travel or teleoperation. Likewise, locomotion interfaces provide a better performance in walking (Patel & Vij, 2010).
The problem with dynamic locomotion is designing hardware as studies in control theory abound, there are comparative fuzziness in the controller’s which involving the mechanical design. The processes and algorithm of the controller cannot directly work on a real world, so they required a transducer or vehicles. Hardware must constantly serve that function by allowing a controllable and efficient platform for locomotion. Clearly, a locomotor will be underwhelming if it’s hardware dynamics are too big or slow for running and walking. Locomotion interfaces also often needed massive hardware which make the installation process difficult. Locomotion is mainly focus on monitoring contact forces, however high proportion gearmotors make force control almost impossible. However, a few of smaller robots have used ultralight legs and direct-drive motor to increase transparency for locomotion. The major reason is force for the robot to move in pipeline with manipulating its locomotion state (Chen, Bai, Lin, & Deng, 2016). Since locomotion is the combination of configuring robot’s pose of and exciting contact forces, the “locomotion-level” orders can be deliver to an interface layer which has been extracts from the exact design. A number of researches has been made on locomotion interface which work as operational-space, whole body and task-space control. Such interfaces focus on force commands and acceleration as the centroid or other parts of the robot and assume the joint torques needed to make accelerations are given friction cones, ground contact, and other restraints.
Locomotion interfaces signify a comparatively new area where there are as yet few researchers and system. The thrust of present study may be defined as exploring the design space of possible approaches and design. This interface also represents a new field, and at this early time there are several uncertainties about the best methods. However, locomotion interface plays a big role in shaping nowadays technology and has significantly improve mechanical field with its implementation.
Conclusion and Future Work
This literature review of empirical research has revealed that locomotion interface helps their user in many ways. The usage of locomotion interface includes in medical, entertainment and mechanical fields. It helps people with stroke rehabilitation, making the entertainment more interesting where user able to experience the virtual world through the VR technologies and helps in manufactured technologies. Through the literature review, it opens an insight on the trends of locomotion interface from previous, current and future. The world future trends are the virtual reality visualization in 3D scenarios which is the VR technologies. VR technologies have a great goal where it can deceive the five human senses where user can believe they are in a real environment yet there are still have problems and limitations appeared in VR technologies. Despite that, we can be sure that future studies comprehensively investigate future trends that provide more realistic experience and immersion in locomotion interface technologies to make a better world for us.
2018-12-17-1545022460