A well-defined deformation model can be vital for non-rigid structure from motion (NRSfM). Most existing methods restrict the deformation space by assuming a fixed rank or smooth deformation, which are not exactly true in the real world, and they require the degree of deformation to be predetermined, which is impractical. Meanwhile, the errors in rotation estimation can have severe effects on the performance, i.e., these errors can make a rigid motion be misinterpreted as a deformation. In this paper, we propose an alternative to resolve these issues, motivated by an observation that non-rigid deformations, excluding rigid changes, can be concisely represented in a linear subspace without imposing any strong constraints, such as smoothness or low-rank. This observation is embedded in our new prior distribution, the Procrustean normal distribution (PND), which is a shape distribution exclusively for non-rigid deformations. Because of this unique characteristic of the PND, rigid and non-rigid changes can be strictly separated, which leads to better performance. The proposed algorithm, EM-PND, fits a PND to given 2D observations to solve NRSfM without any user-determined parameters. The experimental results show that EM-PND gives the state-of-the-art performance for the benchmark data sets, confirming the adequacy of the new deformation model.

In this paper, we propose and demonstrate a novel wireless camera network system, called CITRIC. The core component of this system is a new hardware platform that integrates a camera, a frequency-scalable (up to 624 MHz) CPU, 16 MB FLASH, and 64 MB RAM onto a single device. The device then connects with a standard sensor network mote to form a camera mote. The design enables in-network processing of images to reduce communication requirements, which has traditionally been high in existing camera networks with centralized processing. We also propose a back-end client/server architecture to provide a user interface to the system and support further centralized processing for higher-level applications. Our camera mote enables a wider variety of distributed pattern recognition applications than traditional platforms because it provides more computing power and tighter integration of physical components while still consuming relatively little power. Furthermore, the mote easily integrates with existing low-bandwidth sensor networks because it can communicate over the IEEE 802.15.4 protocol with other sensor network platforms. We demonstrate our system on three applications: image compression, target tracking, and camera localization.

This paper presents a novel class of resource-constrained multi-agent systems for cooperatively estimating an unknown field of interest and locating peaks of the field. Each agent is resource constrained and has limited capabilities in terms of sensing, computation, and communication; hence a centralized approach is not desirable and not feasible. We propose an algorithm for distributed learning and cooperative control of a multi-agent system so that a global goal of the overall system is achieved from locally acting agents. The proposed algorithm is executed by each agent independently to estimate an unknown field of interest from noisy measurements and to coordinate multiple agents in a distributed manner to discover peaks of the unknown field. Each mobile agent maintains its own local estimate of the field and updates the estimate cooperatively using its own measurements and measurements from nearby agents. Then each agent moves towards peaks of the field using the gradient of its estimated field. Agents are coordinated using a distributed rule so that they avoid collision while maintaining communication connectivity. The propose algorithm is based on a recursive spatial estimation of an unknown field of interest using noisy measurements. We show that the closed-loop dynamics of the proposed multi-agent system can be transformed into a form of a stochastic approximation algorithm and prove its convergence using the Ljung's ordinary differential equation (ODE) approach. Our theoretical results are also verified in simulation.

No abstract

Abstract-This paper considers the problem of pursuit evasion games (PEGs), where a group of pursuers is required to chase and capture a group of evaders in minimum time with the aid of a sensor network. We assume that a sensor network is previously deployed and provides global observability of the surveillance region, allowing an optimal pursuit policy. While sensor networks provide global observability, they cannot provide high quality measurements in a timely manner due to packet losses, communication delays, and false detections. This has been the main challenge in developing a real-time control system using sensor networks. We address this challenge by developing a real-time hierarchical control system which decouples the estimation of evader states from the control of pursuers via multiple layers of data fusion. While a sensor network generates noisy, inconsistent, and bursty measurements, the multiple layers of data fusion convert them into consistent and high quality measurements and forward them to the controllers of pursuers in a timely manner. For this control system, three new algorithms are developed: multi-sensor fusion, multi-target tracking and multi-agent coordination algorithms. The multi-sensor fusion algorithm converts correlated sensor measurements into position estimates, the multi-target tracking algorithm tracks an unknown number of targets, and the multi-agent coordination algorithm coordinates pursuers to capture all evaders in minimum time using a robust minimum-time feedback controller. The combined system is evaluated in simulation and tested in a sensor network deployment. To our knowledge, this paper presents the first demonstration of multi-target tracking using a sensor network without relying on classification.

In this paper, a sparse Markov decision process (MDP) with novel causal sparse Tsallis entropy regularization is proposed. The proposed policy regularization induces a sparse and multi-modal optimal policy distribution of a sparse MDP. The full mathematical analysis of the proposed sparse MDP is provided. We first analyze the optimality condition of a sparse MDP. Then, we propose a sparse value iteration method which solves a sparse MDP and then prove the convergence and optimality of sparse value iteration using the Banach fixed point theorem. The proposed sparse MDP is compared to soft MDPs which utilize causal entropy regularization. We show that the performance error of a sparse MDP has a constant bound, while the error of a soft MDP increases logarithmically with respect to the number of actions, where this performance error is caused by the introduced regularization term. In experiments, we apply sparse MDPs to reinforcement learning problems. The proposed method outperforms existing methods in terms of the convergence speed and performance.

No abstract

In this paper, we propose an uncertainty-aware learning from demonstration method by presenting a novel uncertainty estimation method utilizing a mixture density network appropriate for modeling complex and noisy human behaviors. The proposed uncertainty acquisition can be done with a single forward path without Monte Carlo sampling and is suitable for real-time robotics applications. Then, we show that it can be can be decomposed into explained variance and unexplained variance where the connections between aleatoric and epistemic uncertainties are addressed. The properties of the proposed uncertainty measure are analyzed through three different synthetic examples, absence of data, heavy measurement noise, and composition of functions scenarios. We show that each case can be distinguished using the proposed uncertainty measure and presented an uncertainty-aware learning from demonstration method of an autonomous driving using this property. The proposed uncertainty-aware learning from demonstration method outperforms other compared methods in terms of safety using a complex real-world driving dataset.