RoboCup Rescue project

Tadokoro, Satoshı; Kitano, Hiroaki; Takahashi, Tomoichi; Noda, Itsuki; Matsubara, Hitoshi; Shinjoh, Atsushi; Koto, T.; Takeuchi, Ikuo; Takahashi, Hironao; Matsuno, Fumitoshi; Hatayama, Michinori; Ohta, Masayuki; Tayama, Masayuki; Matsui, Takeshi; Kaneda, Toshiyuki; Chiba, Ryosuke; Takeuchi, Kazumasa A.; Nobe, Jun; Noguchi, Kazuhiko; Kuwata, Y.

doi:10.1163/156855300741807

Cited by 8 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Tele-operated rescue robots are developed for search-and-rescue missions within unknown disordered regions such as collapsed buildings, post-earthquake debris, and natural disaster sites [Tadokoro et al 2000]. However, by only observing the robot's camera image without an efficient human interface system, the operator easily misses the position and direction of the robot, which in turn reduces the probability of achieving critical mission objectives.…”

Section: Resultsmentioning

confidence: 99%

Time follower's vision

Sugimoto

Kagotani

Nii

et al. 2004

ACM SIGGRAPH 2004 Emerging Technologies on - SIGGRAPH '04

Self Cite

View full text Add to dashboard Cite

IntroductionTime Follower's Vision is an innovative visual presentation system for remote vehicle control. Autonomous robots provide a wide variety of enhancements and improvements to daily life, and research of this class of robot continues to advance worldwide. However, for fields like medical treatment, search-and-rescue missions, and interpersonal communication, the optimal approach is generally a robot with advanced movement capabilities but a non-autonomous control mechanism.A non-autonomous type robot requires an operator in control, so an efficient human interface system is essential for good performance. For robotic systems controlled via telexistence, the operator performs remote tasks dexterously with the physical feeling of existing in a surrogate robot working in the remote environment. Although efficient operation systems enabling the operator to sense the remote environment have been developed, they require large-scale and high-cost equipment [Tachi 1998]. Further, operators require extensive training to adequately presume the posture of the vehicle from limited information.Time Follower's Vision is a control system that solves these problems by producing a virtual image using Mixed Reality technology, and presenting the surrounding environment of the vehicle and the vehicle's status to the operator. Therefore, even for inexperienced operators, the posture of the vehicle and the surrounding situation can be readily understood. ExpositionTime Follower's Vision is a vision presentation that captures the size, position and environment of the vehicle, which allows even inexperienced operators to control it with ease. This is achieved by using a simple camera and presenting the captured image onto a device like a computer monitor, where the operator controls the robot by simply looking at the image presented on the screen. Figure 1 shows a vehicle equipped with a camera controlled with Time Follower's Vision. Figure 1: A vehicle operated remotely Figure 2: A snapshot and system outlineThrough this technique, the image captured by the vehicle's camera during remote-control operation is stored in a database along with time and position information. The system searches for the optimal image data within the database, which provides information on both the current position and environment of the vehicle. The image is selected by an evaluation function considering the field of view, the position of the camera, and the current position of the vehicle. After an image has been selected, a CG model of the vehicle is mapped with the chosen image, and a viewpoint is presented to give the operator the impression of actually existing behind the vehicle. Finally, the human interface is developed to allow virtually anyone to perform the control with ease by viewing both the vehicle and its environment. Figure 2 shows a snapshot of an actual system and system outline. The operator can correctly understand the relation of the position of the vehicle and the surrounding environment. Using this system, even if the vehicle ...

show abstract

Section: Resultsmentioning

confidence: 99%

Time follower's vision

Sugimoto

Kagotani

Nii

et al. 2004

ACM SIGGRAPH 2004 Emerging Technologies on - SIGGRAPH '04

Self Cite

View full text Add to dashboard Cite

show abstract

“…The predicted location and detection confidence of each person hypothesis is infered by temporal integration of sensor measurements according to the filtering procedure described in Sec. 4. In contrast to single frame evaluation, the detection performance is reported for the whole series of measurements contained in the dataset.…”

Section: Methodsmentioning

confidence: 99%

“…In urban search and rescue (USAR), reliable robots have to provide a semantically meaningful interpretation of objects within a scene (e.g. victims in collapsed buildings) [4]. In unconstrained environments -as is the case in USAR scenarios -relying only on one type of sensor is often insufficient, while fusing complementary information (i.e.…”

Section: Introductionmentioning

confidence: 99%

A Semantic World Model for Urban Search and Rescue Based on Heterogeneous Sensors

Meyer

Schnitzspan

Kohlbrecher

et al. 2011

RoboCup 2010: Robot Soccer World Cup XIV

View full text Add to dashboard Cite

In urban search and rescue scenarios, typical applications of robots include autonomous exploration of possibly dangerous sites, and the recognition of victims and other objects of interest. In complex scenarios, relying on only one type of sensor is often misleading, while using complementary sensors frequently helps improving the performance. To that end, we propose a probabilistic world model that leverages information from heterogeneous sensors and integrates semantic attributes. This method of reasoning about complementary information is shown to be advantageous, yielding increased reliability compared to considering all sensors separately. We report results from several experiments with a wheeled USAR robot in a complex indoor scenario. The robot is able to learn an accurate map, and to detect real persons and signs of hazardous materials based on inertial sensing, odometry, a laser range finder, visual detection, and thermal imaging. The results show that combining heterogeneous sensor information increases the detection performance, and that semantic attributes can be successfully integrated into the world model.

show abstract

“…Instead of human first responders, there may be situations when rescue robots such as snakes and other types of robots may be useful for searching areas that cannot be directly accessed by human searchers. With the increasing prevalence of toxic material produced through terrorist activity or simply released by the collapse of a building, it has been predicted that rescue robots may play and ever more important role in dealing with the aftermath of a disaster [26][27][28][29][30][31][32].…”

Section: Important Of Urban Search and Rescuementioning

confidence: 99%

Rapid GAIT adoption for snake-like locomotion (RGASL)

Tran¹

2021

Preprint

View full text Add to dashboard Cite

Snake-like robots are low centre of gravity because they are limbless, they also have slender bodies composed of multiple actuating segments. Because of these features, snake robots are widely considered to be most adaptable among all land-based mobile robots. The multi-segmented body that provides their defining characteristic, adaptivity, also brings about the quandary of controlling many actuating segments simultaneously to create directd locomotion. Various methods for snake robot locomotion have been proposed for relatively smooth and flat surfaces. Currently there is no snake robot designed or locomotion method capable of resolving the directed mobility problem in situations where the snake robot is stuck at an impasse, or when it encounters disjointed terrains. There is no method to rapidly create new locomotion that addresses the problem or extensive time delay. This thesis makes the contribution of a modular snake robot called Striker and an elegant solution to create new snake-like robot locomotion on-the-fly, called the Explicit Gait Training (EGT) method. The EGT method allows trainer(s) to rapidly train new kinds of locomotion to address any situation at hand using their knowledge, experiences or even trial and error. The third contribution is the Standard Mobility for Snake Robots (SMMSR) is proposed as a standard platform to evaluate the evvectiveness of snake robot locomotion.

show abstract

RoboCup Rescue project

Cited by 8 publications

References 0 publications

Time follower's vision

Time follower's vision

A Semantic World Model for Urban Search and Rescue Based on Heterogeneous Sensors

Rapid GAIT adoption for snake-like locomotion (RGASL)

Contact Info

Product

Resources

About