Stereo Camera Head Research Articles

A probabilistic framework for object search with 6-DOF pose estimation

This article presents a systematic approach to the problem of autonomous 3D object search in indoor environments, using a two-wheeled non-holonomic robot equipped with an actuated stereo-camera head and processing done on a single laptop. A probabilistic grid-based map encodes the likelihood of object existence in each cell and is updated after each sensing action. The updating schema incorporates characteristic parameters modeled after the robot’s sensing modalities and allows for sequential updating via Bayesian recursion methods. Two types of sensing modalities are used to update the map: a coarse search method (global search) based on a color histogram approach, and a more refined search method (local search) based on Scale-Invariant Feature Transform (SIFT) feature matching. If the local search correctly locates the desired object, its 6-DOF pose is estimated using stereo applied to each SIFT feature (i.e. 3D SIFT feature), which is then fed as measurements into an Extended Kalman Filter (EKF) for sustained tracking. If the local search fails to locate the desired object in a particular cell, the cell is updated in the probability map and the next peak probability cell is identified and planned to using a separate grid-based costmap populated via obstacle detection from stereo, with planning done using an A* planner. Experimental results obtained from the use of this method on a mobile robot are presented to illustrate and validate the approach, confirming that the search strategy can be carried out with modest computation on a single laptop.

Vision-based integrated system for object inspection and handling

Image-based effector servoing is a process of perception–action cycles for handling a robot effector under continual visual feedback. This paper applies visual servoing mechanisms not only for handling objects, but also for camera calibration and object inspection. A 6-DOF manipulator and a stereo camera head are mounted on separate platforms and are steered independently. In a first phase (calibration phase), camera features are determined like the optical axes and the fields of sharp view. In the second phase (inspection phase), the robot hand carries an object into the field of view of one camera, then approaches the object along the optical axis to the camera, rotates the object for reaching an optimal view, and finally the object shape is inspected in detail. In the third phase (assembly phase), the system localizes a board containing holes of different shapes, determines the hole which fits most appropriate to the object shape, then approaches and arranges the object appropriately. The final object insertion is based on haptic sensors, but is not treated in the paper. At present, the robot system has the competence to handle cylindrical and cuboid pegs. For handling other object categories the system can be extended with more sophisticated strategies of the inspection and/or assembly phase.

Self Calibration of the Fixation Movement of a Stereo Camera Head

In this article we show how an active stereo camera head can be made to autonomously learn to fixate objects in space. During fixatio n, the system performs an initial and a correction saccade. In the learning phase the correction saccade is controlled by a crude prewired algorithm, in analogy to a mechanism surmised to exist in the brainstem. A vector-based neural network serves as the adaptive component in our system. A self-organizing fovea improves dramatically the convergence of the learning algorithm and the accuracy of the fixation. As a possible application we describe the visuo-motor coordination of the camera head with an anthropomorphic robot arm.

Centering peripheral features in an indoor environment using a binocular log-polar 4 DOF camera head

Abstract This paper describes a hardware/software system that successfully detects certain stationary objects in an indoor scene. The system uses a four degree of freedom (4 DOF) pan-tilt-verge stereo camera head (built in-house) coupled with an inexpensive hardware image processor. A log-polar (LP) image mapping facilitates the detection of floor-wall-ceiling boundaries. The mapping also substantially reduces the data bandwidth of the input imagery by transforming a video resolution image into 64 × 64 pixel multiresolution image. A model matching algorithm detects the wanted object in the LP domain. If the object is in the low-resolution periphery of the LP image, the camera must be moved to center the high-resolution fovea on the object where stereo-from-vergeance can be used to calculate the distance to the object.

The adaptive control of a four-degrees-of-freedom stereo camera head

The paper describes the use of biologically plausible neural network architectures to address some of the issues associated with the use of stereopsis under variable camera geometry. We report an implementation of a layered (subsumption) architecture for the adaptive control of microsaccadic tracking, and show experimental results demonstrating the use of lattice filter predictors for trajectory modelling. A rather simple, but seemingly adequate, neural network architecture for representing high-dimensional surface approximations ( piluts) is evaluated as a method of encoding the predictive stereo mapping of the ground plane for different head positions.