Road following is an important skill vital to the development and deployment of autonomous vehicles. Over past few decades, a large number road computer vision systems have been developed. All these limitations in their capabilities, arising from assumptions idealized conditions. The show dependency on highly structured roads, homogeneity, simplified shapes, lighting In real world, are only effective specialized cases. This paper proposes system that capable dealing with many limitations,...

As part of the Army's Demo III project, a sensor-based system has been developed to identify roads and enable mobile robot drive along them. A ladar sensor, which produces range images, color camera are used in conjunction locate road surface its boundaries. Sensing is constantly update an internal world model surface. The predict future position focus attention sensors on relevant regions their respective images. also determines most suitable algorithm for locating tracking features images...

Abstract The Defense Applied Research Projects Agency (DARPA) Learning to Ground Vehicles (LAGR) program aims develop algorithms for autonomous vehicle navigation that learn how operate in complex terrain. Over many years, the National Institute of Standards and Technology (NIST) has developed a reference model control system architecture called 4D/RCS been applied kinds robot control, including control. For LAGR program, NIST embedded learning into controller enable small used navigate...

The home lift, position, and rehabilitation (HLPR) chair has a unique design novel capabilities when compared with conventional powered wheelchairs. In addition to mobility, it provides lift can transfer patients. Even though medical devices are developing at rapid pace today, an aspect that is often overlooked in these developments adherence “rider safety standards.” contributions of this paper threefold: (i) system, (ii) experiments results toward improved stability test designs include...

In an outdoor, off-road mobile robotics environment, it is important to identify objects that can affect the vehicle's ability traverse its planned path, and determine their three-dimensional characteristics. paper, a combination of three elements used accomplish this task. An imaging ladar collects range images scene. A color camera, whose position relative known, gather images. Information extracted from these sensors build world model, representation current state world. The model...

The National Institute of Standards and Technology (NIST) has been studying pallet visualization for the automated guided vehicle (AGV) industry. Through a cooperative research development agreement with Transbotics, an AGV manufacturer, NIST developed advanced sensor processing world modeling algorithms to verify location orientation respect AGV. Sensor utilizes two onboard AGV, single scan-line, laser-range units. "Safety" is safety unit located at base forktruck "Panner" panning...

This paper describes a world model that combines variety of sensed inputs and priori information is used to generate on-road off-road autonomous driving behaviors. The system designed in accordance with the principles 4D/RCS architecture. hierarchical, resolution scope at each level minimize computational resource requirements support planning functions for control hierarchy. sensory processing populates fuses from multiple sensors extracts feature information, such as terrain elevation,...

We describe a project to collect and disseminate sensor data for autonomous mobility research. Our goals are provide of known accuracy precision researchers developers enable algorithms be developed using realistically difficult sensory data. This enables quantitative comparisons by running them on the same data, allows groups that lack equipment participate in research, speeds technology transfer providing industry with metrics comparing algorithm performance. Data collected NIST High...

In this paper we demonstrate the use of a dynamic, six-degree-of-freedom (6DOF) laser tracker to empirically evaluate performance real-time visual servoing implementation, with objective establishing general method for evaluating 6DOF dimensional measurements. The provides highly accurate ground truth reference measurements position and orientation an object under motion, can be used as standard calibration evaluation robot control algorithms. implementation in study was developed at Purdue...

A robotic vehicle needs to understand the terrain and features around it if is be able navigate complex environments such as road systems. By taking advantage of fact that vehicles also need accurate knowledge their own location orientation, we have developed a sensing object recognition system based on information about area where expected operate. The collected through aerial surveys, from maps, by previous traverses vehicle. It takes form elevation information, feature (roads, signs,...

The National Institute of Standards and Technology's (NIST) Intelligent Systems Division (ISD) is a participant in the Defense Advanced Research Projects Agency (DARPA) LAGR (Learning Applied to Ground Robots) Program. NIST team's objective for Program embed learning algorithms i nto modules that make up 4D/RCS (Four Dimensional/Real-Time Control System), standard reference model architecture ISD has applied many intelligent systems. This enables vehicle learn navigate complex, off-road...

The Army Research Laboratory (ARL) Robotics Collaborative Technology Alliance (CTA) conducted an assessment and evaluation of multiple algorithms for real-time detection pedestrians in Laser Detection Ranging (LADAR) video sensor data taken from a moving platform. were developed by CTA members then assessed field experiments jointly the National Institute Standards (NIST) ARL. A robust, accurate independent pedestrian tracking system was to provide ground truth. truth used evaluate member...

The best reconstructive strategy for upper lip defects is still in debate. purpose of this study was to analyze the decisions made by international microsurgeons, who were participated through online questionnaire, distributed email and social media network.A case a two-thirds oncologic defect presented via an questionnaire 402 microsurgeons replied their treatment options. data then analyzed according geographic area, microsurgical fellowship, seniority, subspecialty. All using SPSS 22.A...

In order to construct algorithmic solutions the problem of geolocalization user-generated photographs and videos, one must first populate a geographic database with different types objects markings that are likely be seen in videos. Toward end, this paper presents framework for detecting labeling satellite imagery. The we interested characterized by low-level features exist mostly at or beyond limits spatial resolution images. To deal challenges posed extraction such from imagery, confine...

The Home Lift, Position and Rehabilitation (HLPR) Chair, developed at the National Institute of Standards Technology (NIST) within Intelligent Systems Division (ISD) from 2005 to 2006, is currently being evaluated for static stability using computer aided design. HLPR Chair has a unique shape as compared conventional wheelchairs powered chairs. Therefore, evaluation its necessary ensure rider safety. Also, was originally designed built with only manual controls. Recently, been transitioned...

This paper describes a novel Home Lift, Position, and Rehabilitation (HLPR) Chair, designed at National Institute of Standards Technology (NIST), to provide independent patient mobility for indoor tasks, such as moving placing person on toilet or bed, lift assistance accessing kitchen other tall shelves. These functionalities are currently out reach most wheelchair users. One the design motivations HLPR Chair is reduce back injury, typically, an important issue in care this group. The being...

In this paper, we apply two fundamental approaches toward evaluating a static, vision based, six-degree-of-freedom (6DoF) pose determination system that measures the position and orientation of part. The first approach uses groundtruth carefully obtained from laser tracker second doesn't use any external groundtruth. evaluation procedure focuses on characterizing both system's accuracy precision as well effect object viewpoints.