Bayesian estimation for autonomous object manipulation based on tactile sensors

Petrovskaya, Anna; Khatib, O.; Thrun, Sebastian; Ng, Andrew Y.

doi:10.1109/robot.2006.1641793

Cited by 102 publications

(75 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Force-sensitive fingers have been used to control the robot's position [6], i.e., to continuously keep the finger in physical contact while moving the object. It has also been shown that tactile sensors can be used to estimate the 3D pose of objects with known shapes [16]. Notably, little information is recovered from the tactile sensor in this work, resulting in multimodal distributions due to ambiguities during the first grasps, which is a problem we are also dealing with in our work.…”

Section: Related Workmentioning

confidence: 94%

Object identification with tactile sensors using bag-of-features

Schneider

Sturm

Stachniss

et al. 2009

2009 IEEE/RSJ International Conference on Intelligent Robots and Systems

188

158

View full text Add to dashboard Cite

Abstract-In this paper, we present a novel approach for identifying objects using touch sensors installed in the finger tips of a manipulation robot. Our approach operates on low-resolution intensity images that are obtained when the robot grasps an object. We apply a bag-of-words approach for object identification. By means of unsupervised clustering on training data, our approach learns a vocabulary from tactile observations which is used to generate a histogram codebook. The histogram codebook models distributions over the vocabulary and is the core identification mechanism. As the objects are larger than the sensor, the robot typically needs multiple grasp actions at different positions to uniquely identify an object. To reduce the number of required grasp actions, we apply a decision-theoretic framework that minimizes the entropy of the probabilistic belief about the type of the object. In our experiments carried out with various industrial and household objects, we demonstrate that our approach is able to discriminate between a large set of objects. We furthermore show that using our approach, a robot is able to distinguish visually similar objects that have different elasticity properties by using only the information from the touch sensor.

show abstract

Section: Related Workmentioning

confidence: 94%

Object identification with tactile sensors using bag-of-features

Schneider

Sturm

Stachniss

et al. 2009

2009 IEEE/RSJ International Conference on Intelligent Robots and Systems

188

158

View full text Add to dashboard Cite

show abstract

“…To improve performance, we turn to Scaling Series, a method first proposed in Petrovskaya et al (2006) for a tactile localization application. In that application the number of parameters was also too large to perform an importance sampling step in real time in conditions of global uncertainty.…”

Section: Scaling Seriesmentioning

confidence: 99%

“…They proposed the Scaling Series algorithm to efficiently produce a much more informed proposal distribution, one that is concentrated around the areas of high probability mass. We refer the reader to Petrovskaya et al (2006) for details on Scaling Series, but briefly, the algorithm works by performing a series of successive refinements, generating an increasingly informative proposal distribution at each step of the series. The successive refinements are performed by gradually annealing the measurement model from artificially relaxed to realistic.…”

Section: Scaling Seriesmentioning

confidence: 99%

Model based vehicle detection and tracking for autonomous urban driving

2009

View full text Add to dashboard Cite

Situational awareness is crucial for autonomous driving in urban environments. This paper describes the moving vehicle detection and tracking module that we developed for our autonomous driving robot Junior. The robot won second place in the Urban Grand Challenge, an autonomous driving race organized by the U.S. Government in 2007. The module provides reliable detection and tracking of moving vehicles from a high-speed moving platform using laser range finders. Our approach models both dynamic and geometric properties of the tracked vehicles and estimates them using a single Bayes filter per vehicle. We present the notion of motion evidence, which allows us to overcome the low signal-to-noise ratio that arises during rapid detection of moving vehicles in noisy urban environments. Furthermore, we show how to build consistent and efficient 2D representations out of 3D range data and how to detect poorly visible black vehicles. Experimental validation includes the most challenging conditions presented at the Urban Grand Challenge as well as other urban settings.

show abstract

“…However, for highly accurate sensors, even such an adaptive technique might require a huge number of samples in order to achieve a sufficiently high particle density during global localization. Alternatively, one can artificially inflate the measurement uncertainty to achieve a regularization of the likelihood function, e.g., see the Scaling Series approach presented by Petrovskaya et al [11]. Also Kalman filters have limitations in highly non-linear systems and in the case of multi-modal likelihood functions.…”

Section: Related Workmentioning

confidence: 99%

Gaussian mixture models for probabilistic localization

Pfaff

Plagemann

Burgard

2008

2008 IEEE International Conference on Robotics and Automation

View full text Add to dashboard Cite

Abstract-One of the key tasks during the realization of probabilistic approaches to localization is the design of a proper sensor model, that calculates the likelihood of a measurement given the current pose of the vehicle and the map of the environment. In the past, range sensors have become popular for mobile robot localization since they directly measure distance. However, in situations in which the robot operates close to edges of obstacles or in highly cluttered environments, small changes in the pose of the robot can lead to large variations in the acquired range scans. If the sensor model used does not appropriately characterize the resulting fluctuations, the performance of probabilistic approaches may substantially degrade. A common solution is to artificially smooth the likelihood function or to only integrate a small fraction of the measurements. In this paper we present a more fundamental and robust approach which uses mixtures of Gaussians to model the likelihood function for single range measurements. In practical experiments we compare our approach to previous methods and demonstrate that it yields a substantially increase in robustness.

show abstract

Bayesian estimation for autonomous object manipulation based on tactile sensors

Cited by 102 publications

References 8 publications

Object identification with tactile sensors using bag-of-features

Object identification with tactile sensors using bag-of-features

Model based vehicle detection and tracking for autonomous urban driving

Gaussian mixture models for probabilistic localization

Contact Info

Product

Resources

About