Different objectives and paradigms exist for tracking multiple targets when measurements do not contain information about the target identities (IDs). The Symmetric Measurement Equation (SME) filter can be used when o...
Different objectives and paradigms exist for tracking multiple targets when measurements do not contain information about the target identities (IDs). The Symmetric Measurement Equation (SME) filter can be used when one is agnostic to the labels and does not attempt to assign different IDs to the different targets. We present an extension of the Kernel-SME filter that, unlike the original variant, uses adaptive kernel widths that depend on the respective uncertainty. In our evaluation, it outperformed existing SMEbased approaches, while it is only second to a more complex global nearest neighbor tracker.
With the growing availability of high-resolution sensors, processing more than one detection per target becomes increasingly critical when tracking multiple extended objects. However, contemporary sensors often genera...
With the growing availability of high-resolution sensors, processing more than one detection per target becomes increasingly critical when tracking multiple extended objects. However, contemporary sensors often generate spurious detections that need to be considered. Naively employing standard multitarget trackers may result in poor tracking performance for multitarget–multidetection tracking in cluttered environments, and the relevant extensions are nontrivial. This paper introduces a version of the kernel symmetric measurement equation (SME) filter that considers both multidetections and clutter. For a simulated scenario, our novel filter achieved a higher accuracy than the global nearest neighbor (GNN) and a fast variant of the joint probabilistic data association filter (JPDAF).
We present DQV-SLAM (Dual Quaternion Visual SLAM). This novel feature-based stereo visual SLAM framework uses a stochastic filter based on the unscented transform and a progressive Bayes update, avoiding linearization...
详细信息
ISBN:
(数字)9780996452786
ISBN:
(纸本)9781728118406
We present DQV-SLAM (Dual Quaternion Visual SLAM). This novel feature-based stereo visual SLAM framework uses a stochastic filter based on the unscented transform and a progressive Bayes update, avoiding linearization of the nonlinear spatial transformation group. 6-DoF poses are represented by dual quaternions where rotational and translational components are stochastically modeled by Bingham and Gaussian distributions. Maps represented by point clouds of ORB-features are incrementally built and landmarks are updated with an unscented transform-based method. In order to get reliable measurements during the update, an optical flow-based approach is proposed to remove false feature associations. Drift is corrected by pose graph optimization once loop closure is detected. The KITTI and EuRoC datasets for stereo setup are used for evaluation. The performance of the proposed system is comparable to state-of-the-art optimization-based SLAM systems and better than existing filtering-based approaches.
Serious climate changes and energy-related environmental problems are currently critical issues in the world. In order to reduce carbon emissions and save our environment, renewable energy harvesting technologies will...
详细信息
Acquiring high resolution images in deep regions is challenging in ultrasound imaging due to limited probe aperture size and low transmit frequency usage. The concept of synthetic tracked aperture ultrasound (STRATUS)...
详细信息
ISBN:
(纸本)9781467398985
Acquiring high resolution images in deep regions is challenging in ultrasound imaging due to limited probe aperture size and low transmit frequency usage. The concept of synthetic tracked aperture ultrasound (STRATUS) imaging is introduced to extend the effective aperture size by moving the probe while accurately tracking its orientation and translation. Based on the synthetic aperture technique, sub-apertures from each pose can be synthesized to construct a high-resolution image. In particular, we propose a mechanical tracking configuration using a 6 degree-of-freedom (DOF) robotic arm with force sensors that not only provides a robust tracking accuracy, but also enables co-operative control. The ultrasound probe is moved by an operator, while a virtual fixture uses force feedback of the robotic arm to constrain the motion to be on a desired plane or trajectory. Furthermore, we developed an algorithm to mitigate the potential errors between consecutive poses, such as tracking inaccuracy, tissue deformation, and phase aberration. Those errors were extracted by computing subtle image shift through cross-correlation for all neighboring poses, and the procedure is dynamically applied to the entire image. Comparing the STRATUS image to a conventional single pose image, the full width at the half maximum (FWHM) of a point target located at a depth of around 85 mm improved from 3.13 mm to 2.78 mm, and SNR improved from 28.96 dB to 30.27 dB. In addition, the dynamic error compensation further improved the FWHM and SNR to be 1.15 mm and 33.17 dB, respectively. The results proved the feasibility of the co-robotic STRATUS imaging, and dynamic error compensation improved the system's tolerance to errors.
The accuracy of the Autonomous Underwater Vehicles (AUVs) navigation system determines whether they can safely operate and return. Traditional Dead-reckoning (DR) relies on the inertial sensors such as gyroscope and a...
详细信息
ISBN:
(纸本)9781450346375
The accuracy of the Autonomous Underwater Vehicles (AUVs) navigation system determines whether they can safely operate and return. Traditional Dead-reckoning (DR) relies on the inertial sensors such as gyroscope and accelerometer. A major challenge for DR navigation is from measurement error of the inertial sensors (gyroscope, accelerometer, etc.), especially when the AUV is near or at the ocean surface. The AUV's motion is affected by ocean waves, and its pitch angle changes rapidly with the waves. This rapid change and the measurement errors will cause great noise to the direction measured by gyroscopes, and then lead to a large error to the DR navigation. To address this problem, a novel DR method based on neural network (DR-N) is proposed to explore the time-varying relationship between acceleration measurement and orientation measurement, which leverages acoustic localization and neural network estimate timely pitch angle through the explored time-varying relationship. This method enables AUV's DR navigation with a single acceleration, without relying on both acceleration and gyroscope. Most importantly, we can improve the accuracy of AUV navigation through avoiding DR errors caused by gyroscope noise at the sea surface. Simulations show DR-N significantly improves navigation accuracy.
State estimation concepts like the Kalman filter heavily rely on potentially noisy sensor data. In general, the estimation quality depends on the amount of sensor data that can be exploited. However, missing observati...
详细信息
ISBN:
(纸本)9781509020126
State estimation concepts like the Kalman filter heavily rely on potentially noisy sensor data. In general, the estimation quality depends on the amount of sensor data that can be exploited. However, missing observations do not necessarily impair the estimation quality but may also convey exploitable information on the system state. This type of information-noted as negative information-often requires specific measurement and noise models in order to take advantage of it. In this paper, a hybrid Kalman filter concept is employed that allows using both stochastic and set-membership representations of information. In particular, the latter representation is intended to account for negative information, which can often be easily described as a bounded set in the measurement space. Depending on the type of information, the filtering step of the proposed estimator adaptively switches between Gaussian and ellipsoidal noise representations. A target tracking scenario is studied to evaluate and discuss the proposed concept.
This paper presents a simulator for swarm operations designed to verify algorithms for a swarm of autonomous underwater robots (AUVs), specifically for constructing an underwater communication network with AUVs carryi...
详细信息
ISBN:
(纸本)9781450346375
This paper presents a simulator for swarm operations designed to verify algorithms for a swarm of autonomous underwater robots (AUVs), specifically for constructing an underwater communication network with AUVs carrying acoustic communication devices. This simulator consists of three nodes: a virtual vehicle node (VV), a virtual environment node (VE), and a visual showing node (VS). The modular design treats AUV models as a combination of virtual equipment. An expert acoustic communication simulator is embedded in this simulator, to simulate scenarios with dynamic acoustic communication nodes. The several simulations we have performed demonstrate that this simulator is easy to use and can be further improved.
This paper introduces the concept of proactive execution of robot tasks in the context of human-robot cooperation with uncertain knowledge of the human's intentions. We present a system architecture that defines t...
详细信息
This paper introduces the concept of proactive execution of robot tasks in the context of human-robot cooperation with uncertain knowledge of the human's intentions. We present a system architecture that defines the necessary modules of the robot and their interactions with each other. The two key modules are the intention recognition that determines the human user's intentions and the planner that executes the appropriate tasks based on those intentions. We show how planning conflicts due to the uncertainty of the intention information are resolved by proactive execution of the corresponding task that optimally reduces the system's uncertainly. Finally, we present an algorithm for selecting this task and suggest a benchmark scenario.
暂无评论