Autonomous navigation and mapping using monocular low-resolution grayscale vision

Murali, Vidya N.; Birchfield, Stan

doi:10.1109/cvprw.2008.4563136

Cited by 13 publications

(9 citation statements)

References 53 publications

(35 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After describing these individual measures, we present an integrated system that is able to autonomously explore an unknown indoor environment, recovering from difficult situations like corners, blank walls, and initial heading toward a wall. This approach extends our previous work [29], in which reactive behavior was demonstrated with a subset of measures, but without explicit estimation of orientation or distance to the end. All of this behavior is accomplished at a rate of 1000 Hz on a standard computer using only 0.02% of the pixels available from a standard 30 Hz color VGA (640 × 480) video camera, discarding 99.98% of the information.…”

Section: Introductionsupporting

confidence: 50%

“…The only input to the system consisted of the 32 × 24 downsampled grayscale images from the 30 Hz camera. In our previous work we estimated orientation using only the median of bright pixels, and distance to the end of the corridor was largely determined by entropy [29,27]. In this work we show that a linear combination (weighted average) of five (orientation) and three (distance to the end) complementary measures is more effective for achieving success in multiple environments.…”

Section: Exploration In An Unknown Environmentmentioning

confidence: 93%

See 1 more Smart Citation

Autonomous exploration using rapid perception of low-resolution image information

Murali

Birchfield

2011

Auton Robot

Self Cite

View full text Add to dashboard Cite

We present a technique for mobile robot exploration in unknown indoor environments using only a single forward-facing camera. Rather than processing all the data, the method intermittently examines only small 32 × 24 downsampled grayscale images. We show that for the task of indoor exploration the visual information is highly redundant, allowing successful navigation even using only a small fraction of the available data. The method keeps the robot centered in the corridor by estimating two state parameters: the orientation within the corridor, and the distance to the end of the corridor. The orientation is determined by combining the results of five complementary measures, while the estimated distance to the end combines the results of three complementary measures. These measures, which are predominantly information-theoretic, are analyzed independently, and the combined system is tested in several unknown corridor buildings exhibiting a wide variety of appearances, showing the sufficiency of low-resolution visual information for mobile robot exploration. Because the algorithm discards such a large percentage of the pixels both spatially and temporally, processing occurs at an average of 1000 frames per second, thus freeing the processor for other concurrent tasks.

show abstract

Section: Introductionsupporting

confidence: 50%

Section: Exploration In An Unknown Environmentmentioning

confidence: 93%

Autonomous exploration using rapid perception of low-resolution image information

Murali

Birchfield

2011

Auton Robot

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recent self-guided vehicles used in the DARPA LAGR programme have led to significant advances in robotic perception systems (Hadsell et al, 2009;Sofman et al, 2006;Kim et al, 2007), however the multiple sensors and complexity of these systems do not address the needs of low-cost autonomous robots (Katramados et al, 2009;Murali and Birchfield, 2008). A commonly available web camera presents a desirable alternative that will be used in this work, and is motivated by the human ability to interpret 2D low resolution images (Murali and Birchfield, 2008).…”

Section: Introductionmentioning

confidence: 99%

A Generative Traversability Model for Monocular Robot Self-guidance

Sapienza¹,

Camilleri²

2012

Proceedings of the 9th International Conference on Informatics in Control, Automation and Robotics

View full text Add to dashboard Cite

Abstract:In order for robots to be integrated into human active spaces and perform useful tasks, they must be capable of discriminating between traversable surfaces and obstacle regions in their surrounding environment. In this work, a principled semi-supervised (EM) framework is presented for the detection of traversable image regions for use on a low-cost monocular mobile robot. We propose a novel generative model for the occurrence of traversability cues, which are a measure of dissimilarity between safe-window and image superpixel features. Our classification results on both indoor and outdoor images sequences demonstrate its generality and adaptability to multiple environments through the online learning of an exponential mixture model. We show that this appearance-based vision framework is robust and can quickly and accurately estimate the probabilistic traversability of an image using no temporal information. Moreover, the reduction in safe-window size as compared to the state-of-the-art enables a self-guided monocular robot to roam in closer proximity of obstacles.

show abstract

“…Similarly, sensory-motor learning has been used to map visual inputs to turning commands, but the resulting algorithms have been too computationally demanding for real-time performance (Giovannangeli et al, 2006). Other researchers have developed mapless algorithms for low-level functionality like corridor following or obstacle avoidance (Nelson & Aloimonos, 1988, Santos-Victor et al, 1995, Barrows et al, 2002, LeCun et al, 2005, Michels et al, 2005, Murali & Birchfield, 2008, but these techniques are not applicable to following a specific arbitrary path. The open circle coincides with both the camera focal point and the robot position, the arrow indicates the heading direction,  is the image plane, and  is the angle between the optical axis and the projection ray from the landmark.…”

mentioning

confidence: 99%