Abstract

Field Programmable Gate Arrays (FPGA) exceed the computing power of software based implementations by breaking the paradigm of sequential execution and accomplishing more per clock cycle by enabling hardware level parallelization at an architectural level. Introducing parallel architectures for a computationally intensive algorithm like Rapidly Exploring Random Trees (RRT) will result in an exploration that is fast, dense and uniform. Through a cost function delineated in later sections, FPGA based combinatorial architecture delivers superlative speed-up but consumes very high power while hierarchical architecture delivers relatively lower speed-up with acceptable power consumption levels. To combine the qualities of both, a hybrid architecture, that encompasses both combinatorial and hierarchical architecture, is designed. To determine the number of RRT nodes to be allotted to the combinatorial and hierarchical blocks of the hybrid architecture, a cost function, comprised of fundamentally inversely related speed-up and power parameters, is formulated. This maximization of cost function, with its associated constraints, is then mathematically solved using a modified branch and bound, that leads to optimal allocation of RRT modules to both blocks. It is observed that this hybrid architecture delivers the highest performance-per-watt out of the three architectures for differential, quad-copter and fixed wing kinematics.

Abstract

In this paper, we present a framework for generating smooth and stable trajectories for wheeled mobile robots moving on uneven terrains. Instead of relying on static stability measures, the paper incorporates velocity and acceleration based constraints like no-slip and permanent wheel ground contact conditions in the planning framework. The paper solves this complicated problem in a computationally efficient manner with full 3D dynamics of the robot. The two major aspects of the proposed work are: Firstly, closed form functional relationships are derived which describes the 6 dof evolution of the robot’s state on an arbitrary 2.5D uneven terrain. This enables us to have a fast evaluation of robot’s dynamics along any candidate trajectory. Secondly, a novel concept of non-linear time scaling is introduced through which simple algebraic relations defining the bounds on velocities and accelerations are obtained. Moreover, non-linear time scaling also provides a new approach for manipulating velocities and accelerations along given geometric paths. It is thus exploited to obtain stable velocity and acceleration profiles. Extensive simulation results are presented to demonstrate the efficacy of the proposed methodology.

Abstract

Curb detection is a critical component of driver assistance and autonomous driving systems. In this paper, we present a discriminative approach to the problem of curb detection under diverse road conditions. We define curbs as the intersection of drivable and non-drivable area which are classified using dense Conditional random fields(CRF). In our method, we fuse output of a neural network used for pixel-wise semantic segmentation with depth and color information from stereo cameras. CRF fuses the output of a deep model and height information available in stereo data and provides improved segmentation. Further we introduce temporal smoothness using a weighted average of Segnet output and output from a probabilistic voxel grid as our unary potential. Finally, we show improvements over the current state of the art neural networks. Our proposed method shows accurate results over large range of variations in curb curvature and appearance, without the need of retraining the model for the specific dataset.

Abstract

Frontier detection is a critical component in autonomous exploration, wherein the robot decides the next best location to move in order to continue its mapping process. The existing frontier detection methods require dense reconstruction which is difficult to attain in a poorly textured indoor environment using a monocular camera. In this effort, we present an alternate method of detecting frontiers during the course of robot motion that circumvents the requirement of dense mapping. Based on the observation that frontiers typically occur around areas with sudden change in texture (zero-crossings), we propose a novel linear chain Conditional Random Field(CRF) formulation that is able to detect the presence or absence of frontier regions around such areas. We use cues like spread of 3D points and scene change around these areas as an observation to CRF. We demonstrate that this method gives us more relevant frontiers compared to other monocular camera based methods in the literature. Finally, we present results in an indoor environment, wherein frontiers are reliably detected around walls leading to new corridors, doors leading to new rooms or corridors and tables and other objects that open up to a new space in rooms.

Abstract

Recently, the interest in Micro Aerial Vehicles (MAVs) and their autonomous flights has increased tremendously and significant advances have been made. The monocular camera has turned out to be most popular sensing modality for MAVs as it is light-weight, does not consume more power, and encodes rich information about the environment around. In this paper, we present DeepFly, our framework for autonomous navigation of a quadcopter equipped with monocular camera. The navigable space detection and waypoint selection are fundamental components of autonomous navigation system. They have broader meaning than just detecting and avoiding immediate obstacles. Finding the navigable space emphasizes equally on avoiding obstacles and detecting ideal regions to move next to. The ideal region can be defined by two properties: 1) All the points in the region have approximately same high depth value and 2) The area covered by the points of the region in the disparity map is considerably large. The waypoints selected from these navigable spaces assure collision-free path which is safer than path obtained from other waypoint selection methods which do not consider neighboring information. In our approach, we obtain a dense disparity map by performing a translation maneuver. This disparity map is input to a deep neural network which predicts bounding boxes for multiple navigable regions. Our deep convolutional neural network with shortcut connections regresses variable number of outputs without any complex architectural add on. Our autonomous navigation approach has been successfully tested in both indoors and outdoors environment and in range of lighting conditions.

Abstract

Autonomous navigation of generic monocular quadcopter in the indoor environment requires sophisticated approaches for perception, planning and control. This paper presents a system which enables a miniature quadcopter with a frontal monocular camera to autonomously navigate and explore the unknown indoor environment. Initially, the system estimates dense depth map of the environment from a single video frame using our proposed novel supervised Hierarchical Structured Learning (hsl) technique, which yields both high accuracy levels and better generalization. The proposed hsl approach discretizes the overall depth range into multiple sets. It structures these sets hierarchically and recursively through partitioning the set of classes into two subsets with subsets representing apportioned depth range of the parent set, forming a binary tree. The binary classification method is applied to each internal node of binary tree separately using Support Vector Machine (svm). Whereas, the depth estimation of each pixel of the image starts from the root node in top-down approach, classifying repetitively till it reaches any of the leaf node representing its estimated depth. The generated depth map is provided as an input to Convolutional Neural Network (cnn), which generates flight planning commands. Finally, trajectory planning and control module employs a convex programming technique to generate collision-free minimum time trajectory which follows these flight planning commands and produces appropriate control inputs for the quadcopter. The results convey unequivocally the advantages of depth perception by hsl, while repeatable flights of successful nature in typical indoor corridors confirm the efficacy of the pipeline.

Abstract

Crowded urban environments are composed of different types of dynamic and static elements. Learning and classification of features is a major task in solving the localization problem in such environments. This work presents a gradual learning methodology to learn the useful features using multiple experiences. The usefulness of an observed element is evaluated by a scoring mechanism which uses two scores – reliability and distinctiveness. The visual features thus learned are used to partition the visual map into smaller regions. The robot is efficiently localized in such a partitioned environment using two-level localization. The concept of active map (AM) is proposed here, which is a map that represents one partition of the environment in which there is a high probability of the robot existing. High-level localization is used to track the mode of the AMs using discrete Bayes filter. Low-level localization uses a bag-of …

Abstract

The Simultaneous Localization and Mapping problem (SLAM) in robotics is typically modeled as a dyadic graph of relative pose measurements taken by the robot. The graph nodes store the values representing the absolute pose of the robot at a given point of time. An edge connecting two nodes represents robot movement and it stores the measurements taken by the robot sensor while moving between two nodes. The objective of the SLAM problem is to find the optimal global measurements best satisfying the noisy relative measurements [12]. This problem of optimal estimation on a graph given relative measurements is a well-studied problem within the control community, for which several results and algorithms are known [3, 4]. SLAM is generally solved as a least squares problem. Robust kernels which are less sensitive to outliers are used to deal with noise and outlier measurements. However, robust kernels tend to be dependent on initialization and can fail as the number of outliers increase. Therefore, it's important to identify and prune the outlier (noisy) measurements represented by incorrect loop closure edges for an accurate pose estimate. In this paper we propose a multi-scale Heat-Kernel analysis based loop closure edge pruning algorithm for the SLAM graph. We show that compared to other pruning algorithms, our algorithm has a substantially higher precision and recall when compared and is able to handle a large amount of outlier measurements.We have corroborated results on several publicly available datasets and several types of noise. Our algorithm is not restricted to SLAM graphs only, but has a much wider applicability to other types of geometric graphs.

Abstract

Complex tasks are often handled through software implementation in combination with high performance processors. Taking advantage of hardware parallelism, FPGA is breaking the paradigm by accomplishing more per clock cycle with closely matched application requirements. With the aim to minimise computation delay with increase in map's size and geometric constraints, we present the FPGA based combinatorial architecture that allows multiple RRTs to work together to achieve accelerated, uniform exploration of the map. We also analyse our architecture against hardware implementation of other scalable RRT methods for motion planning. We observe notable furtherance of acceleration capabilities with the proposed architecture delivering a minimum 3X gain over the other implementations while maintaining uniformity in exploration

Abstract

In an object search scenario with several small objects spread over a large indoor environment, the robot cannot infer about all of them at once. Pruning the search space is highly desirable in such a case. It has to actively select a course of actions to closely examine a selected set of objects. Here, we model the inferences about far away objects and their viewpoint priors into a decision analytic abstraction to prioritize the waypoints. By selecting objects of interest, a potential field is built over the environment by using Composite Viewpoint Object Potential (CVOP) maps. A CVOP is built using VOP, a framework to identify discriminative viewpoints to recognize small objects having distinctive features only in specific views. Also, a CVOP helps to clearly disambiguate objects which look similar from far away. We formulate a Decision Analysis Graph (DAG) over the above information, to assist the robot in actively navigating and maximize the reward earned. This optimal strategy increases search reliability, even in the presence of similar looking small objects which induce confusion into the agent and simultaneously reduces both time taken and distance travelled. To the best of our knowledge, there is no current unified formulation which addresses indoor object search scenarios in this manner. We evaluate our system over ROS using a TurtleBot mounted with a Kinect.