SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
This paper discusses the Team Raptor passive perception system hardware and software design, implementation, and performance, and describes a road map to faster and improved passive perception.
Explore the source record for details and available documents.
Detecting water hazards is a significant challenge to unmanned ground vehicle autonomous off-road navigation. This paper focuses on detecting the presence of water during the daytime using color cameras. A multi-cue approach is taken. Evidence of the presence of water is generated from color, texture, and the detection of reflections in stereo range data. A rule base for fusing water cues was developed by evaluating detection results from an extensive archive of data collection imagery containing water. This software has been implemented into a run-time passive perception subsystem and tested thus far under Linux on a Pentium based processor.
The classical notion of how things are seen is that perception is passive, that the eyes are windows, and in floods reality. Physiological work of the 19th century cast doubt on this view that perception is passive acceptance of reality. Perception is not at the present time a popular topic for philosophers. This must be partly because scientific accounts of perception have now gone a long way away from appearances. They depend on physiological and psycho-physical experiments which require technical investigation and do not fall within traditional concepts of philosophy. Theories of visual perception are examined, both from a physical and psycho-physical standpoint.
The ability to perform off-road autonomous navigation at any time of day or night is a requirement for some unmanned ground vehicle (UGV) programs. Because there are times when it is desirable for military UGVs to operate without emitting strong, detectable electromagnetic signals, a passive only terrain perception mode of operation is also often a requirement. Thermal infrared (TIR) cameras can be used to provide day and night passive terrain perception. TIR cameras have a detector sensitive to either mid-wave infrared (MWIR) radiation (3-5?m) or long-wave infrared (LWIR) radiation (8-12?m). With the recent emergence of high-quality uncooled LWIR cameras, TIR cameras have become viable passive perception options for some UGV programs. The Jet Propulsion Laboratory (JPL) has used a stereo pair of TIR cameras under several UGV programs to perform stereo ranging, terrain mapping, tree-trunk detection, pedestrian detection, negative obstacle detection, and water detection based on object reflections. In addition, we have evaluated stereo range data at a variety of UGV speeds, evaluated dual-band TIR classification of soil, vegetation, and rock terrain types, analyzed 24 hour water and 12 hour mud TIR imagery, and analyzed TIR imagery for hazard detection through smoke. Since TIR cameras do not currently provide the resolution available from megapixel color cameras, a UGV's daytime safe speed is often reduced when using TIR instead of color cameras. In this paper, we summarize the UGV terrain perception work JPL has performed with TIR cameras over the last decade and describe a calibration target developed by General Dynamics Robotic Systems (GDRS) for TIR cameras and other sensors.
TIR cameras can be used for day/night Unmanned Ground Vehicle (UGV) autonomous navigation when stealth is required. The quality of uncooled TIR cameras has significantly improved over the last decade, making them a viable option at low speed Limiting factors for stereo ranging with uncooled LWIR cameras are image blur and low texture scenes TIR perception capabilities JPL has explored includes: (1) single and dual band TIR terrain classification (2) obstacle detection (pedestrian, vehicle, tree trunks, ditches, and water) (3) perception thru obscurants
Successful off-road autonomous navigation by an unmanned ground vehicle (UGV) requires reliable perception and representation of natural terrain. While perception algorithms are used to detect driving hazards, terrain mapping algorithms are used to represent the detected hazards in a world model a UGV can use to plan safe paths. There are two primary ways to detect driving hazards with perception sensors mounted to a UGV: binary obstacle detection and traversability cost analysis. Binary obstacle detectors label terrain as either traversable or non-traversable, whereas, traversability cost analysis assigns a cost to driving over a discrete patch of terrain. In uncluttered environments where the non-obstacle terrain is equally traversable, binary obstacle detection is sufficient. However, in cluttered environments, some form of traversability cost analysis is necessary. The Jet Propulsion Laboratory (JPL) has explored both approaches using stereo vision systems. A set of binary detectors has been implemented that detect positive obstacles, negative obstacles, tree trunks, tree lines, excessive slope, low overhangs, and water bodies. A compact terrain map is built from each frame of stereo images. The mapping algorithm labels cells that contain obstacles as no-go regions, and encodes terrain elevation, terrain classification, terrain roughness, traversability cost, and a confidence value. The single frame maps are merged into a world map where temporal filtering is applied. In previous papers, we have described our perception algorithms that perform binary obstacle detection. In this paper, we summarize the terrain mapping capabilities that JPL has implemented during several UGV programs over the last decade and discuss some challenges to building terrain maps with stereo range data.
Water detection is a critical perception requirement for unmanned ground vehicle (UGV) autonomous navigation over cross-country terrain. During the Robotics Collaborative Technology Alliances (RCTA) program, the Jet Propulsion Laboratory (JPL) developed a set of water detection algorithms that are used to detect, localize, and avoid water bodies large enough to be a hazard to a UGV. The JPL water detection software performs the detection and localization stages using a forward-looking stereo pair of color cameras. The 3D coordinates of water body surface points are then output to a UGV's autonomous mobility system, which is responsible for planning and executing safe paths. There are three primary methods for evaluating the performance of the water detection software. Evaluations can be performed in image space on the intermediate detection product, in map space on the final localized product, or during autonomous navigation to characterize the avoidance of a variety of water bodies. This paper describes a methodology for performing the first two types of water detection performance evaluations.
Robust water detection is a critical perception requirement for unmanned ground vehicle (UGV) autonomous navigation. This is particularly true in wide-open areas where water can collect in naturally occurring terrain depressions during periods of heavy precipitation and form large water bodies. One of the properties of water useful for detecting it is that its surface acts as a horizontal mirror at large incidence angles. Water bodies can be indirectly detected by detecting reflections of the sky below the horizon in color imagery. The Jet Propulsion Laboratory (JPL) has implemented a water detector based on sky reflections that geometrically locates the pixel in the sky that is reflecting on a candidate water pixel on the ground and predicts if the ground pixel is water based on color similarity and local terrain features. This software detects water bodies in wide-open areas on cross-country terrain at mid- to far-range using imagery acquired from a forward-looking stereo pair of color cameras mounted on a terrestrial UGV. In three test sequences approaching a pond under a clear, overcast, and cloudy sky, the true positive detection rate was 100% when the UGV was beyond 7 meters of the water's leading edge and the largest false positive detection rate was 0.58%. The sky reflection based water detector has been integrated on an experimental unmanned vehicle and field tested at Ft. Indiantown Gap, PA, USA.
Reliable detection of non-traversable hazards is a key requirement for off-road autonomous navigation. A detailed description of each obstacle detection algorithm and their performance on the surveyed obstacle course is presented in this paper.
Adaptive changes during space flight in how the brain integrates vestibular cues with other sensory information can lead to impaired movement coordination and spatial disorientation following G-transitions. This ESA-NASA study was designed to examine both the physiological basis and operational implications for disorientation and tilt-translation disturbances following short-duration spaceflights. The goals of this study were to (1) examine the effects of stimulus frequency on adaptive changes in motion perception during passive tilt and translation motion, (2) quantify decrements in manual control of tilt motion, and (3) evaluate vibrotactile feedback as a sensorimotor countermeasure.
This joint ESA-NASA study is examining changes in motion perception following Space Shuttle flights and the operational implications of post-flight tilt-translation ambiguity for manual control performance. Vibrotactile feedback of tilt orientation is also being evaluated as a countermeasure to improve performance during a closed-loop nulling task. METHODS. Data has been collected on 5 astronaut subjects during 3 preflight sessions and during the first 8 days after Shuttle landings. Variable radius centrifugation (216 deg/s) combined with body translation (12-22 cm, peak-to-peak) is utilized to elicit roll-tilt perception (equivalent to 20 deg, peak-to-peak). A forward-backward moving sled (24-390 cm, peak-to-peak) with or without chair tilting in pitch is utilized to elicit pitch tilt perception (equivalent to 20 deg, peak-to-peak). These combinations are elicited at 0.15, 0.3, and 0.6 Hz for evaluating the effect of motion frequency on tilt-translation ambiguity. In both devices, a closed-loop nulling task is also performed during pseudorandom motion with and without vibrotactile feedback of tilt. All tests are performed in complete darkness. PRELIMINARY RESULTS. Data collection is currently ongoing. Results to date suggest there is a trend for translation motion perception to be increased at the low and medium frequencies on landing day compared to pre-flight. Manual control performance is improved with vibrotactile feedback. DISCUSSION. The results of this study indicate that post-flight recovery of motion perception and manual control performance is complete within 8 days following short-duration space missions. Vibrotactile feedback of tilt improves manual control performance both before and after flight.
Astronauts experience sensorimotor changes during spaceflight, particularly during G-transitions. Post flight sensorimotor changes include spatial disorientation, along with postural and gait instability that may degrade operational capabilities of the astronauts and endanger the crew. A sensorimotor countermeasure that mitigates these effects would improve crewmember safety and decrease risk. The goal of this research is to investigate the potential use of stochastic vestibular stimulation (SVS) as a technology to improve sensorimotor function. We hypothesize that low levels of SVS will improve sensorimotor perception through the phenomenon of stochastic resonance (SR), when the response of a nonlinear system to a weak input signal is enhanced by the application of a particular nonzero level of noise. This study aims to advance the development of SVS as a potential countermeasure by 1) demonstrating the exhibition of stochastic resonance in vestibular perception, a vital component of sensorimotor function, 2) investigating the repeatability of SR exhibition, and 3) determining the relative contribution of the semicircular canals (SCC) and otolith (OTO) organs to vestibular perceptual SR. A constant current stimulator was used to deliver bilateral bipolar SVS via electrodes placed on each of the mastoid processes, as previously done. Vestibular perceptual motion recognition thresholds were measured using a 6-degree of freedom MOOG platform and a 150 trial 3-down/1-up staircase procedure. In the first test session, we measured vestibular perceptual thresholds in upright roll-tilt at 0.2 Hz (SCC+OTO) with SVS ranging from 0-700 μA. In a second test session a week later, we re-measured roll-tilt thresholds with 0, optimal (from test session 1), and 1500 μA SVS levels. A subset of these subjects, plus naive subjects, participated in two additional test sessions in which we measured thresholds in supine roll-rotation at 0.2 Hz (SCC) and upright y-translation at 1 Hz (OTO) with SVS up to 700 μA. A sinusoidal galvanic vestibular stimulation (GVS) perceptual threshold was also measured on each test day and used to normalize the SVS levels across subjects. In roll-tilt thresholds with SVS, the characteristic SR curve was qualitatively exhibited in 10 of 12 subjects, and the improvement in motion threshold was significant in 6 subjects, indicating that optimal SVS improved passive body motion perception in a way that is consistent with classical SR theory. A probabilistic comparison to numeric simulations further validated these experimental results. On the second test session, 4 out of the 10 SR exhibitors showed repeated improvement with SVS compared to the no SVS condition. Data collection is ongoing for the last two test sessions in which SCC and OTO only perceptual motion recognition thresholds are being measured with SVS. The final results of these test sessions will give insight into whether vestibular perceptual SR can occur when only one type of vestibular sensor is sensing motion or if it is more evident when sensory integration between the SCC and OTO is occurring during the motion. The overall purpose of this research is to further quantify the effects of SVS on various sensorimotor tasks and to gain a more fundamental understanding of how SVS causes SR in the vestibular system. In the context of human space flight, results from this research will help in understanding how SVS may be practically implemented in the future as a component of a comprehensive countermeasure plan for G-transition adaptation.
In many passive visual tasks, human perceptual judgments are contrast dependent. To explore whether these contrast dependencies of visual perception also affect closed-loop manual control tasks, we examined visuomotor performance as humans actively controlled a moving luminance-defined line over a range of contrasts. Four subjects were asked to use a joystick to keep a horizontal line centered on a display as its vertical position was perturbed by a sum of sinusoids under two control regimes. The total root mean square (RMS) position error decreased quasi-linearly with increasing log contrast across the tested range (mean slope across subjects: -8.0 and -7.7% per log2 contrast, for the two control regimes, respectively). Frequency-response (Bode) plots showed a systematic increase in open-loop gain (mean slope: 1.44 and 1.30 dB per log2 contrast, respectively), and decrease in phase lag with increasing contrast, which can be accounted for by a decrease in response time delay (mean slope: 32 and 40 ms per log2 contrast, respectively). The performance data are well fit by a Crossover Model proposed by McRuer and Krendel, which allowed us to identify both visual position and motion cues driving performance. This analysis revealed that the position and motion cues used to support manual control under both control regimes appear equally sensitive to changes in stimulus contrast. In conclusion, our data show that active control of a moving visual stimulus is as dependent on contrast as passive perception and suggest that this effect is attributed to a shared contrast sensitivity early in the visual pathway, before any specialization for motion processing.
Four astronauts experienced passive whole-body rotation in a number of test sessions during a 7-day orbital mission. Pitch (Y-axis) and roll (X-axis) rotation required subject orientations on the rotator in which the otolith system was at radius of 0.5 m. Thus subjects experienced a constant -0.22 Gz stimulus to the otoliths during the 60 s constant-velocity segments of "pitch" and "roll" ramp profiles. The Gz stimulus, a radius-dependent vector ranging from -0.22 Gz at the otoliths to +0.36 Gz at the feet, generated sensory information that was not interpreted as inversion in any of the 16 tests carried out in flight (12 in pitch and 4 in roll orientation). None of the subjects was rotated with head off-center during the first 33 h of the mission. In the state of orbital adaptation of these subjects, a -0.22 Gz otolith stimulus did not provide a vertical reference in the presence of a gradient of +Gz stimuli to the trunk and legs.
Motion sickness typically is considered a bothersome artifact of exposure to passive motion in vehicles of conveyance. This condition seldom has significant impact on the health of individuals because it is of brief duration, it usually can be prevented by simply avoiding the eliciting condition and, when the conditions that produce it are unavoidable, sickness dissipates with continued exposure. The studies conducted examined several aspects of motion sickness in animal models. A principle objective of these studies was to investigate the neuroanatomy that is important in motion sickness with the objectives of examining both the utility of putative models and defining neural mechanisms that are important in motion sickness.
Psychophysical research has demonstrated that human observers utilize a variety of visual cues to form a perception of three-dimensional depth. However, most of these studies have utilized a passive judgement paradigm, and failed to consider depth-cue integration as a dynamic and task-specific process. In the current study, we developed and experimentally validated a model of manual control of depth that examines how two potential cues (stereo disparity and relative size) are utilized in both first- and second-order active depth control tasks. We found that stereo disparity plays the dominate role for determining depth position, while relative size dominates perception of depth velocity. Stereo disparity also plays a reduced role when made less salient (i.e., when viewing distance is increased). Manual control models predict that position information is sufficient for first-order control tasks, while velocity information is required to perform a second-order control task. Thus, the rules for depth-cue integration in active control tasks are dependent on both task demands and cue quality.
The present conference discusses topics in the fusion of active and passive sensors, object estimation and verification, three-dimensional representation and knowledge integration, three-dimensional perception from multisensor data, the representation of uncertainty in multisensor fusion, and sensor calibration and registration. Also discussed are the areas of multisensor target detection and classification, multisensor processing architectures, knowledge structures and spatial reasoning, sensory interfaces to telerobotic systems, and navigation with spatial data bases.