A5073562330- Planetary Science and Exploration
- Robotics and Sensor-Based Localization
- Robotic Path Planning Algorithms
- Space Exploration and Technology
- Modular Robots and Swarm Intelligence
- Soil Mechanics and Vehicle Dynamics
- Robotic Locomotion and Control
- Astro and Planetary Science
- Geochemistry and Geologic Mapping
- Space Science and Extraterrestrial Life
- Space Satellite Systems and Control
- Advanced Image and Video Retrieval Techniques
- Underwater Vehicles and Communication Systems
- Spaceflight effects on biology
- Spacecraft Design and Technology
- Distributed systems and fault tolerance
- Target Tracking and Data Fusion in Sensor Networks
- Remote-Sensing Image Classification
- Advanced Vision and Imaging
- Polar Research and Ecology
- Video Surveillance and Tracking Methods
- Remote Sensing and LiDAR Applications
- Machine Learning and Algorithms
- Microbial Community Ecology and Physiology
- Isotope Analysis in Ecology
Carnegie Mellon University
2015-2024
Carnegie Mellon University Qatar
2011
Canadian Space Agency
2011
Robotics Research (United States)
2007-2009
University of Iowa
2006
Ames Research Center
1998-2002
Australian National University
1999-2002
Dante II is a unique walking robot that provides important insight into high-mobility robotic locomotion and remote exploration. II’s uniqueness stems from its combined legged rappelling mobility system, scanning-laser rangefinder, multilevel control scheme. In 1994 was deployed successfully tested in Alaskan volcano, as demonstration of the fieldworthiness these technologies. For more than five days explored alone volcano crater using combination supervised autonomous teleoperated control....
Abstract We describe a simultaneous localization and mapping (SLAM) method for hovering underwater vehicle that will explore caves tunnels, true three‐dimensional (3D) environment. Our consists of Rao‐Blackwellized particle filter with 3D evidence grid map representation. procedure dynamically adjusting the number particles to provide real‐time performance. also how we adjust prediction step accommodate sensor degradation or failure. present an efficient octree data structure makes it...
In a two-year study of collaborative human-robot system, researchers observed science team in Pittsburgh and robot Chile. The system was part project intended to inform planetary exploration while studying terrestrial desert. Over two years, autonomy increased such that the could execute sequences commands and, under certain circumstances, make autonomous decisions about instrument deployment data collection. Analysis observational artifact addressed impact increasing levels on how users...
Scarab is a prototype rover for Lunar missions to survey resources in polar craters. It designed as prospector that would use deep coring drill and apply soil analysis instruments measure the abundance of elements hydrogen oxygen other volatiles including water. Scarab’s chassis can adjust wheelbase height stabilize its contact with ground also posture better ascend descend steep slopes. This enables unique control when moving introduces new planning issues. has undergone field testing at...
Sediments in the hyper-arid core of Atacama Desert are a terrestrial analog to Mars regolith. Understanding distribution and drivers microbial life sediment may give critical clues on how search for biosignatures Mars. Here, we identify spatial highly specialized bacterial communities previously unexplored depth horizons subsurface sediments 800 mm. We deployed an autonomous rover mission-relevant Martian drilling scenario with manual sample validation. Subsurface were delineated by related...
Automatically correlating plant observable characteristics to their underlying genetics will streamline selection methods in breeding. Measurement of is called phenotyping, and knowing phenotypes accurately throughout a plant's growth central making breeding decisions. In-field phenotyping an automated noninvasive manner hence crucial accelerating methods. However, most the existing on using visual imaging are confined controlled greenhouse environments. This paper presents method mapping 2D...
In this paper, we present an active SLAM framework for volumetric exploration of 3D underwater environments with multibeam sonar. Recent work in integrated and planning performs localization while maintaining free-space information. However, absence informative loop closures can lead to imperfect maps, therefore unsafe behavior. To solve this, propose a navigation policy that reduces vehicle pose uncertainty by balancing between revisitation. identify locations revisit, build visual...
The Life in the Atacama project investigated regional distribution of life and habitats Desert Chile. We sought to create biogeologic maps through survey traverses across desert using a rover carrying biologic geologic instruments. Elements our science approach were to: Perform ecological transects from relatively wet coastal range arid core desert; use converging evidence instruments reach conclusions about microbial abundance; develop test exploration strategies adapted search scattered...
Geologists and planetary scientists will benefit from methods for accurate segmentation of rocks in natural scenes. However, are poorly suited current visual techniques - they exhibit diverse morphologies have no uniform property to distinguish them background soil. We address this challenge with a novel detection method incorporating features multiple scales. These include local attributes such as texture, object shading two-dimensional shape, scene the direction illumination. Our uses...
In order to be deployed in real-world driving environments, self-driving cars must able recognize and respond exceptional road conditions, such as highway workzones, because unusual events can alter previously known traffic rules geometry. this paper, we present a set of computer vision methods that recognize, through identification workzone signs, the bounds temporary changes environments. Through testing using video data about workzones recorded under various weather our approach was...
This work presents an experimental method for visualizing and analyzing machine‐soil interactions, namely the soil optical flow technique (SOFT). SOFT uses clustering techniques to process images of interacting with a wheel or tool from photos taken through glass wall bin. It produces results that are far richer than past approaches utilized long‐exposure photography. achieves performance comparable particle image velocimetry tracking velocimetry, but without need specialized measurement...
Laser Raman spectroscopy (LRS) has been proposed for in situ characterization of molecular species planetary surface exploration, and three laser spectrometers are included the science payloads two under‐development missions to Mars (ESA‐ExoMars2018 NASA‐Mars2020). We report first rover test a spectrometer developed flight, Micro‐beam Spectrometer (MMRS) Atacama Desert (Chile). The MMRS was integrated on Zoë analyzed subsurface samples brought up by 1 m drill delivered carousel. demonstrated...
Atacama Desert, the oldest and driest desert on Earth, displays significant rains only once per decade. To investigate how microbial communities take advantage of these wet events, we carried out a geomicrobiological study few days after heavy rain event in 2015. Different physicochemical community analyses were conducted samples collected from playas an alluvial fan surface, 10, 20, 50 80 cm depth. A water content peaks measured 10 20 depth (from 1.65% to 4.1% w/w maximum values) while,...
This letter addresses the path planning problem for a rover on deformable, sloped terrains. As such terrains induce high slip and possible immobilization of rovers, finding that avoids critical is important traversing However, it difficult to predict slippage precisely, especially steep slopes, certain level prediction uncertainty inevitable. Although several algorithms consider have been proposed thus far, they do not in prediction. proposes robust algorithm finds safe efficient based...
Ahstruct-Gait generation is the formulation and selection of a sequence coordinated leg body motions that propel legged robot along desired path. Approaches to gait can be classified into control, behavioral, rule-based, constraintbased paradigms. We survey these models introduce Ambler, hexapod circulate its legs produce unique gaits. Then we present kinematic, collision, terrain, support, stability constraints available for generation, discuss our progress using constraint-based method...
We present a method for infrastructure-free localization of underwater vehicles with multibeam sonar. After constructing large scale (4 km), high resolution (1 m) bathymetric map region the ocean floor, vehicle can use to correct its gradual dead-reckoning error, or re-localize itself after returning from surface. This ability is particularly important deep-operating vehicles, which accumulate amounts error during descent through water column. 3D evidence grid, stored in efficient octree...
The deep phreatic thermal explorer (DEPTHX) is an autonomous underwater vehicle designed to navigate unexplored environment, generate high-resolution three-dimensional (3-D) maps, collect biological samples based on sampling decision, and return its origin. In the spring of 2007, DEPTHX was deployed in Zacatón, a (∼318 m), limestone, sinkhole (cenote) northeastern Mexico. As descended, it generated 3-D map processing range data from 54 onboard sonars. collected water column wall biomat...