Abstract

In this paper, we address the problem of reactive collision avoidance among multiple robots. However we impose the constraint of continuous curvature on the collision avoidance manoeuvre which adds to the complexity of the problem but is necessary for realistic implementation of collision avoidance among multiple non-holonomic robots like tricycles and car-like. Since these non-holonomic robots cannot execute point-turn motion, performing collision avoidance with these robots while maintaining curvature continuity becomes a very challenging problem. Hence the proposed work comes as an enhancement over the existing reactive collision avoidance frameworks which completely neglect the above aspects. We modify the collision cone approach of collision avoidance to the case of multiple robots moving with non-linearly varying velocities. This is done by approximating the vehicle evolution with a smooth function represented as a linear combination of orthogonal basis functions and writing the collision avoidance condition as analytical function with respect to time. As a direct consequence of that, the reactive collision avoidance and goal-reaching could be solved as a constrained optimization problem. The theory is substantiated by extensive simulations for various bench-mark cases.

Abstract

Over the past couple of years, with the development of efficient control algorithms, micro aerial vehicles have come into the picture. In this paper, we consider the problem of creating a map of an indoor environment which provides more information than a 2D map and at the same time is more accurate than the contemporary 3D mapping algorithms. We propose a novel collaborative system, consisting of an unmanned ground vehicle and a micro aerial vehicle, which is used to create augmented 2D maps using the distinct sensing capabilities of these two robots. This system works in a collaborative manner such that the two robots complement each others movement and sensing capabilities.

Abstract

This paper presents a method to capture the orbiting objects using a robotic system mounted on a service satellite. The main objective is to manipulate the robot such that no reaction moment gets transferred to the base satellite. This will avoid use of any attitude controller resulting in fuel savings. Note that the constraints leading to zero reaction moment are nonholonomic, and this makes path planning a complex problem. In this work, first a method based on holonomic distribution of the nonholonomic constraints is discussed. As this method exploits constraints in terms of joint velocities, it does not always ensure successful capture. Next, a method based on task-level constraints, written in terms of end-effector's velocities, has been illustrated. It is shown that the path planned using this method has several singular points. In order to overcome disadvantages of the above two methods a novel approach is proposed which uses holonomic distribution to reach closer to the target and task-level constraints to finally capture the target. Efficacy of the method is shown using a 3-link robot mounted on a service satellite.

Abstract

Single camera mobile robots, wherein the single camera becomes the quintessential sensor for robotic tasks such as localization, mapping and obstacle avoidance are challenging. From such a standpoint, we demonstrate a dense reconstruction as conducted by a navigating robot with a monocular camera. Unlike most other dense reconstruction methods this approach first identifies planar areas through homography. These segments are tracked over multiple views with homography based dense correspondences. The tracked correspondences are reconstructed within a VSLAM formulation, wherein the dense reconstructed points get added to the existing SLAM computed structure. The dense structure is further refined using a modified Bundle Adjustment which minimizes projection error in 3D to align with a inferred model of the scene. A mobile robot can thus make use of the reconstructed ground plane and planar obstacles to compute a collision free trajectory.

Abstract

In this paper we propose a semi-active robot for climbing steep obstacles like steps, curbs, etc. The key novelty of the proposed robot lies in the use of a passive mechanism for climbing steps of smaller heights and motor only while climbing steps of bigger heights. Analysis of the robot's stability during its ascent phase is also investigated. Model based control is used to achieve step climbing. The other novelty of the robot, in contrast to existing active suspension step climbers, is that it does not need the knowledge of step height beforehand. Therefore, the mechanism has the advantage of height-independent climbing motion as in the case of passive mechanism along with the extra freedom of active joints for maintaining vehicle stability, only when required. Efficacy of the mechanism is exhibited through simulations on steps of various heights.

Abstract

In this paper we develop an algorithm to generate gait sequences to negotiate a discontinuous terrain for a hybrid 4-wheeled legged robot. The gait sequence comprises two main steps - normal force redistribution and hybrid position-force control. The robot climbs the discontinuity one leg at a time. This requires that the entire load of the robot is taken up by the other three legs so that the leg climbing the discontinuity is free. For this purpose a load redistribution methodology is used which makes the center of gravity of chassis coincide with the desired center of pressure (CoP). Subsequently the free leg moves in hybrid position and force control to climb the discontinuity. Force sensing ensures constant contact with the terrain and detection of stand and end of the discontinuity without using any perception sensor. The methodology is validated using multi-body dynamic simulation.

Abstract

This paper proposes a novel method of integrating planning with Monocular Simultaneous Localization and Mapping (SLAM) systems. Monocular SLAM, typically referred to as VSLAM systems in literature consists of recovering trajectory estimates of the camera and stationary world features from a single moving camera. Such VSLAM systems are significantly more difficult than SLAM performed with depth sensors, such as using an accurate Laser Range Finder (LRF). When the camera motion is subject to steep changes in orientation, tracked features over the previous instances are lost, making VSLAM estimates highly unreliable, erroneous that cannot be recovered. Most often a complete breakdown occurs, which entails a new sequence of images to be captured from a fresh camera trajectory. Herein we propose an optimization based path planning formulation for such VSLAM systems that reduces occurence of such errors through paths that are not subject to high orientation changes. Further we plan a velocity profile over the path that prevents features from getting significantly displaced over successive images, often considered a critical criteria for robust feature tracking. The velocity profile is computed using the novel concept of non linear time scaling proposed in our earlier work. The VSLAM system is also sufficiently innovated to provide for dense mapping over planar segments. The efficacy of the formulation is verified over real experiments on a camera mounted robot.

Abstract

Depth information has been shown to affect identification of visually salient regions in images. In this paper, we investigate the role of depth in saliency detection in the presence of (i) competing saliencies due to appearance,(ii) depth-induced blur and (iii) centre-bias. Having established through experiments that depth continues to be a significant contributor to saliency in the presence of these cues, we propose a 3D-saliency formulation that takes into account structural features of objects in an indoor setting to identify regions at salient depth levels. Computed 3D saliency is used in conjunction with 2D saliency models through non-linear regression using SVM to improve saliency maps. Experiments on benchmark datasets containing depth information show that the proposed fusion of 3D saliency with 2D saliency models results in an average improvement in ROC scores of about 9% over state-of-the-art 2D saliency models.

Abstract

Visual localization in crowded dynamic environments requires information about static and dynamic objects. This paper presents a robust method that learns the useful features from multiple runs in highly crowded urban environments. Useful features are identified as distinctive ones that are also reliable to extract in diverse imaging conditions. Relative importance of features is used to derive the weight for each feature. The popular Bag-of-words model is used for image retrieval and localization, where query image is the current view of the environment and database contains the visual experience from previous runs. Based on the reliability, features are augmented and eliminated over runs. This reduces the size of representation, and makes it more reliable in crowded scenes. We tested the proposed method on data sets collected from highly crowded Indian urban outdoor settings. Experiments have shown that with the help of a small subset (10%) of the detected features, we can reliably localize the camera. We achieve superior results in terms of localization accuracy even when more than 90% of the pixels are occluded or dynamic.

Abstract

In this paper we consider the problem of coordinating the motion of the manipulator and the vehicle to produce stable trajectories for the combined mobile manipulator system on uneven terrain. These kinds of situations often arise in planetary exploration, where rovers equipped with a manipulator are required to navigate over general uneven terrain. Moreover the framework can also be used in situations where the mobile manipulator is required to transport objects on uneven terrain. We generate feasible trajectories for the vehicle between a given start and a goal point considering the dynamics of the manipulator. The framework proposed in the paper plans such motion profile of the manipulator that maximizes vehicle stability which is measured by a novel concept called Feasible Acceleration Count (FAC). We show that, from the point of view of motion planning of mobile manipulator on uneven terrains, FAC gives a better estimate of vehicle stability than more popular metrics like Tip-Over Stability. The trajectory planner closely resembles motion primitive based graph based planning and is combined with a novel cost function derived from FAC. The efficacy of the approach is shown through simulations of a mobile manipulator system on a 2.5D uneven terrain.