Christian Siagian

Biologically Inspired Mobile Robot Vision Localization

We present a robot localization system using biologically inspired vision. Our system models two extensively studied human visual capabilities: (1) extracting the ldquogistrdquo of a scene to produce a coarse localization hypothesis and (2) refining it by locating salient landmark points in the scene. Gist is computed here as a holistic statistical signature of the image, thereby yielding abstract scene classification and layout. Saliency is computed as a measure of interest at every image location, which efficiently directs the time-consuming landmark-identification process toward the most likely candidate locations in the image. The gist features and salient regions are then further processed using a Monte Carlo localization algorithm to allow the robot to generate its position. We test the system in three different outdoor environments-building complex (38.4 m times 54.86 m area, 13 966 testing images), vegetation-filled park (82.3 m times 109.73 m area, 26 397 testing images), and open-field park (137.16 m times 178.31 m area, 34 711 testing images)-each with its own challenges. The system is able to localize, on average, within 0.98, 2.63, and 3.46 m, respectively, even with multiple kidnapped-robot instances.

Mobile robot vision navigation & localization using Gist and Saliency

We present a vision-based navigation and localization system using two biologically-inspired scene understanding models which are studied from human visual capabilities: (1) Gist model which captures the holistic characteristics and layout of an image and (2) Saliency model which emulates the visual attention of primates to identify conspicuous regions in the image. Here the localization system utilizes the gist features and salient regions to accurately localize the robot, while the navigation system uses the salient regions to perform visual feedback control to direct its heading and go to a user-provided goal location. We tested the system on our robot, Beobot2.0, in an indoor and outdoor environment with a route length of 36.67m (10,890 video frames) and 138.27m (28,971 frames), respectively. On average, the robot is able to drive within 3.68cm and 8.78cm (respectively) of the center of the lane.

Mobile robot monocular vision navigation based on road region and boundary estimation

We present a monocular vision-based navigation system that incorporates two contrasting approaches: region segmentation that computes the road appearance, and road boundary detection that estimates the road shape. The former approach segments the image into multiple regions, then selects and tracks the most likely road appearance. On the other hand, the latter detects the vanishing point and road boundaries to estimate the shape of the road. Our algorithm operates in urban road settings and requires no training or camera calibration to maximize its adaptability to many environments. We tested our system in 1 indoor and 3 outdoor urban environments using our ground-based robot, Beobot 2.0, for real-time autonomous visual navigation. In 20 trial runs the robot was able to travel autonomously for 98.19% of the total route length of 316.60m.

Biologically-Inspired Face Detection: Non-Brute-Force-Search Approach

We present a biologically-inspired face detection system. The system applies notions such as saliency, gist, and gaze to localize a face without performing blind spatial search. The saliency model consists of highly parallel low-level computations that operate in domains such as intensity, orientation, and color. It is used to direct attention to a set of conspicuous locations in an image as starting points. The gist model, computed in parallel with the saliency model, estimates holistic image characteristics such as dominant contours and magnitude in high and low spatial frequency bands. We are limiting its use to predicting the likely head size based on the entire scene. Also, instead of identifying face as a single entity, this system performs detection by parts and uses spatial configuration constraints to be robust against occlusion and perspective.

Mobile robot navigation system in outdoor pedestrian environment using vision-based road recognition

We present a mobile robot navigation system guided by a novel vision-based road recognition approach. The system represents the road as a set of lines extrapolated from the detected image contour segments. These lines enable the robot to maintain its heading by centering the vanishing point in its field of view, and to correct the long term drift from its original lateral position. We integrate odometry and our visual road recognition system into a grid-based local map that estimates the robot pose as well as its surroundings to generate a movement path. Our road recognition system is able to estimate the road center on a standard dataset with 25,076 images to within 11.42 cm (with respect to roads at least 3 m wide). It outperforms three other state-of-the-art systems. In addition, we extensively test our navigation system in four busy college campus environments using a wheeled robot. Our tests cover more than 5 km of autonomous driving without failure. This demonstrates robustness of the proposed approach against challenges that include occlusion by pedestrians, non-standard complex road markings and shapes, shadows, and miscellaneous obstacle objects.

Beobot 2.0: Cluster architecture for mobile robotics

With the recent proliferation of robust but computationally demanding robotic algorithms, there is now a need for a mobile robot platform equipped with powerful computing facilities. In this paper, we present the design and implementation of Beobot 2.0, an affordable research-level mobile robot equipped with a cluster of 16 2.2-GHz processing cores. Beobot 2.0 uses compact Computer on Module (COM) processors with modest power requirements, thus accommodating various robot design constraints while still satisfying the requirement for computationally intensive algorithms. We discuss issues involved in utilizing multiple COM Express modules on a mobile platform, such as interprocessor communication, power consumption, cooling, and protection from shocks, vibrations, and other environmental hazards such as dust and moisture. We have applied Beobot 2.0 to the following computationally demanding tasks: laser-based robot navigation, scale-invariant feature transform (SIFT) object recognition, finding objects in a cluttered scene using visual saliency, and vision-based localization, wherein the robot has to identify landmarks from a large database of images in a timely manner. For the last task, we tested the localization system in three large-scale outdoor environments, which provide 3,583, 6,006, and 8,823 test frames, respectively. The localization errors for the three environments were 1.26, 2.38, and 4.08 m, respectively. The per-frame processing times were 421.45, 794.31, and 884.74 ms respectively, representing speedup factors of 2.80, 3.00, and 3.58 when compared to a single dual-core computer performing localization.

Comparison of gist models in rapid scene categorization tasks

Discussions and Conclusions The capacity of humans to perform a number of complex visual tasks such as scene categorization in as little as 100ms has been attributed to their ability to rapidly extract the “gist” of a scene [1 -3]. Following a brief presentation of a photograph, an observer is able to summarize the quintessential characteristics of an image, a process previously expected to require much analysis [4-5].

Storing and recalling information for vision localization

In implementing a vision localization system, a crucial issue to consider is how to efficiently store and recall the necessary information so that the robot is not only able to accurately localize itself, but does so in a timely manner. In the presented system, we discuss a strategy to minimize the amount of stored data by analyzing the strengths and weaknesses of several cooperating recognition modules, and by using them through a prioritization scheme, which orders the data entries from the most likely to match to the least. We validate the system is a series of experiments at three large scale outdoor environments: a building complex (126 times 180 ft. area, 3583 testing images), a vegetation-filled park (270 times 360 ft. area, 6006 testing images), and an open-field area (450 times 585 ft. area, 8823 testing images) - each with its own set of challenges. Not only is the system able to localize in these environments (on average 3.46 ft., 6.55 ft. 12.96 ft. of error, respectively), it does so while searching through only 7.35%, 3.50%, and 6.12% of all the stored information, respectively.

Autonomous Mobile Robot Localization and Navigation Using a Hierarchical Map Representation Primarily Guided by Vision

While impressive progress has recently been made with autonomous vehicles, both indoors and on streets, autonomous localization and navigation in less constrained and more dynamic environments, such as outdoor pedestrian and bicycle-friendly sites, remains a challenging problem. We describe a new approach that utilizes several visual perception modules-place recognition, landmark recognition, and road lane detection-supplemented by proximity cues from a planar laser range finder for obstacle avoidance. At the core of our system is a new hybrid topological/grid-occupancy map that integrates the outputs from all perceptual modules, despite different latencies and time scales. Our approach allows for real-time performance through a combination of fast but shallow processing modules that update the maps state while slower but more discriminating modules are still computing. We validated our system using a ground vehicle that autonomously traversed three outdoor routes several times, each 400 m or longer, on a university campus. The routes featured different road types, environmental hazards, moving pedestrians, and service vehicles. In total, the robot logged over 10 km of successful recorded experiments, driving within a median of 1.37 m laterally of the center of the road, and localizing within 0.97 m median longitudinally of its true location along the route.

Fixation bank: Learning to reweight fixation candidates

Predicting where humans will fixate in a scene has many practical applications. Biologically-inspired saliency models decompose visual stimuli into feature maps across multiple scales, and then integrate different feature channels, e.g., in a linear, MAX, or MAP. However, to date there is no universally accepted feature integration mechanism. Here, we propose a new a data-driven solution: We first build a “fixation bank” by mining training samples, which maintains the association between local patterns of activation, in 4 feature channels (color, intensity, orientation, motion) around a given location, and corresponding human fixation density at that location. During testing, we decompose feature maps into blobs, extract local activation patterns around each blob, match those patterns against the fixation bank by group lasso, and determine weights of blobs based on reconstruction errors. Our final saliency map is the weighted sum of all blobs. Our system thus incorporates some amount of spatial and featural context information into the location-dependent weighting mechanism. Tested on two standard data sets (DIEM for training and test, and CRCNS for test only; total of 23,670 training and 15,793 + 4,505 test frames), our model slightly but significantly outperforms 7 state-of-the-art saliency models.