Conference Papers

Learning to Assess the Quality of Stroke Rehabilitation Exercises

Lee M.H., Siewiorek D.P., Smailagic A., Bernadino A.,Bermudez S.

IUI ’19, March 17–20, 2019, Marina del Ray, CA, USA

2018

Abstract:

Due to the limited number of therapists, task-oriented exercises are often prescribed for post-stroke survivors as inhome rehabilitation. During in-home rehabilitation, a patient may become unmotivated or confused to comply prescriptions without the feedback of a therapist. To address this challenge, this paper proposes an automated method that can achieve not only qualitative, but also quantitative assessment of stroke rehabilitation exercises. Specifically, we explored a threshold model that utilizes the outputs of binary classifiers to quantify the correctness of a movements into a performance score. We collected movements of 11 healthy subjects and 15 post-stroke survivors using a Kinect sensor and ground truth scores from primary and secondary therapists. The proposed method achieves the following agreement with the primary therapist: 0.8436, 0.8264, and 0.7976 F1-scores on three task-oriented exercises. Experimental results show that our approach performs equally well or better than multi-class classification, regression, or the evaluation of the secondary therapist. Furthermore, we found a strong correlation (R 2 = 0.95) between the sum of computed exercise scores and the Fugl-Meyer Assessment scores, clinically validated motor impairment index of post-stroke survivors. Our results demonstrate a feasibility of automatically assessing stroke rehabilitation exercises with the decent agreement levels and clinical relevance.

Lessons learned from a real vehicular network deployment of delay-tolerant networking

Monteiro R., Guedes L., Condeixa T., Neves F., Sargento S., Guardalben L., Steenkiste P.

2015 IEEE International Conference on Communication Workshop, ICCW 2015

2015

Abstract:

Vehicular networks are inherently unstable networks with high mobility and intermittent connectivity. These networks can greatly benefit from Delay Tolerant Networking (DTN) solutions for opportunistic connectivity in the transmission of delay-tolerant data. In this paper, we evaluate the performance of different DTN routing protocols in real world vehicular networks with different degrees of connectivity in order to understand their feasibility in real vehicular environments. Our case-study application is the upload of delay-tolerant sensing data from vehicular nodes to a server on the Internet. We deploy DTN in 3 vehicular testbeds: a heterogeneous lab testbed with controlled mobility, a low-mobility high-connectivity network of harbor trucks, and a high-mobility low-connectivity city bus network. We compare 3 routing protocols: epidemic, static and PRoPHET, for which we measure delivery ratio, average delay, path length, as well as transmission and storage overhead. We analyze the results for experimental insight, and extract lessons for DTN routing protocol design.

Leveraging parked cars as urban self-organizing Road-Side Units

Reis A.B., Sargento S.

2015 IEEE 82nd Vehicular Technology Conference, VTC Fall 2015 - Proceedings

2016

Abstract:

In urban vehicular networks, Road-Side Units (RSUs) take a crucial role in improving the performance of the network, by working as content distribution points, relays for time-critical broadcasts, and points of central coordination. The high costs associated with the installation and maintenance of RSUs, however, keep these units from seeing widespread deployment. One approach to this problem is for cars to be used opportunistically as RSUs, and in urban areas, the presence of large numbers of parked cars make these entities promising candidates for establishing vehicular support networks. In this paper we introduce new methods for parked cars to self-organize and act as a support network to the existing urban vehicular network, alleviating the need for costly deployments of fixed road-side units. Our approach considers parked cars that can both complement existing fixed RSUs and take the role of RSUs themselves, improving the network’s performance on multiple applications. We show that even a small number of parked cars can bring considerable improvements to the network, and that our proposed methods for self-organization create support networks of parked cars that can cover the urban area with an optimal numbers of vehicles.

Linear Convergence Rate of a Class of Distributed Augmented Lagrangian Algorithms

Jakovetic D., Moura J.M.F., Xavier J.

IEEE Transactions on Automatic Control

2015

Abstract:

We study distributed optimization where nodes cooperatively minimize the sum of their individual, locally known, convex costs f i (x)’s; x ϵ ℝ d is global. Distributed augmented Lagrangian (AL) methods have good empirical performance on several signal processing and learning applications, but there is limited understanding of their convergence rates and how it depends on the underlying network. This paper establishes globally linear (geometric) convergence rates of a class of deterministic and randomized distributed AL methods, when the f i ‘s are twice continuously differentiable and have a bounded Hessian. We give explicit dependence of the convergence rates on the underlying network parameters. Simulations illustrate our analytical findings.

Linear logical relations for session-based concurrency

Perez J.A., Caires L., Pfenning F., Toninho B.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

2012

Abstract:

In prior work we proposed an interpretation of intuitionistic linear logic propositions as session types for concurrent processes. The type system obtained from the interpretation ensures fundamental properties of session-based typed disciplines—most notably, type preservation, session fidelity, and global progress. In this paper, we complement and strengthen these results by developing a theory of logical relations. Our development is based on, and is remarkably similar to, that for functional languages, extended to an (intuitionistic) linear type structure. A main result is that well-typed processes always terminate (strong normalization). We also introduce a notion of observational equivalence for session-typed processes. As applications, we prove that all proof conversions induced by the logic interpretation actually express observational equivalences, and explain how type isomorphisms resulting from linear logic equivalences are realized by coercions between interface types of session-based concurrent systems.

Localization and Navigation Assessment of a Heavy-Duty Field Robot

Nasir A.K., Araújo A.G., & Couceiro M.S.

IROS 2020 Workshop on Perception, Planning and Mobility in Forestry Robotics (WPPMFR 2020)

2020

Abstract:

Autonomous robots for challenging domains, for instances transportation, agriculture, forestry and construction, face several technical challenges inherent to their ability, or lack of it, in operating under the unstructured, harsh and dynamic nature of outdoor environments. This paper presents preliminary results of system integration toward autonomous navigation of a heavy-duty ground mobile robot designed for such challenging outdoor domains. The main contribution of this paper is to incorporate existing state-of-the-art Robot Operating System (ROS) based algorithms for localization, mapping, traversability, navigation, and exploration in real world unknown and unstructured environments. This paper focuses on assessing the localization and navigation ability of the robot by using proposed methodology and evaluating it under real-world outdoor tests.

Locating cars through a vision enabled VANET

Ferreira M., Conceicao H., Fernandes R., Reis R.

IEEE Intelligent Vehicles Symposium, Proceedings

2009

Abstract:

With the introduction of cameras in production cars, and the wide dissemination of wireless inter-vehicle communication devices in all new vehicles in the foreseeing future, emerging applications of this ldquoomnipresent hardware networkrdquo can be designed. That is the case of the localization process of a particular vehicle, as the result of a broadcasted search warrant from some particular authority. The system here proposed relies in the optical recognition of the license plate of the wanted car. We present some results of license plate recognition based on real-world images acquired in different driving conditions, as well as of simulation of such a distributed search in a large-scale urban VANET, that show the feasibility of the whole process. A cryptographic protocol is proposed to deal with the security issues of the system.

Low-power clock synchronization using electromagnetic energy radiating from AC power lines

Rowe A., Gupta V., Rajkumar R.

Proceedings of the 7th ACM Conference on Embedded Networked Sensor Systems, SenSys 2009

2009

Abstract:

Clock synchronization is highly desirable in many sensor networking applications. It enables event ordering, coordinated actuation, energy-efficient communication and duty cycling. This paper presents a novel low-power hardware module for achieving global clock synchronization by tuning to the magnetic field radiating from existing AC power lines. This signal can be used as a global clock source for battery-operated sensor nodes to eliminate drift between nodes over time even when they are not passing messages. With this scheme, each receiver is frequency-locked with each other, but there is typically a phase-offset between them. Since these phase offsets tend to be constant, a higher-level compensation protocol can be used to globally synchronize a sensor network. We present the design of an LC tank receiver circuit tuned to the AC 60Hz signal which we call a Syntonistor. The Syntonistor incorporates a low-power microcontroller that filters the signal induced from AC power lines generating a pulse-per-second output for easy interfacing with sensor nodes. The hardware consumes less than 58μW which is 2–3 times lower than the idle state of most sensor networking MAC protocols. Next, we evaluate a software clock-recovery technique running on the local microcontroller that minimizes timing jitter and provides robustness to noise. Finally, we provide a protocol that sets a global notion of time by accounting for phase-offsets. We evaluate the synchronization accuracy and energy performance as compared to in-band message passing schemes. The use of out-of-band signals for clock synchronization has the useful property of decoupling the synchronization scheme from any particular MAC protocol. Our experiments show that over a 11 day period, eight nodes distributed across the floor of the CIC building on Carnegie Mellon’s campus remained synchronized on an average to less than 1ms without exchanging any radio messages beyond the initialization phase.

Lynceus: Cost-efficient Tuning and Provisioning of Data Analytic Jobs

Casimiro M., Didona D., Romano P., Rodrigues L., Zwanepoel W., Garlan D.

ICDCS’2020

2020

Abstract:

Modern data analytic and machine learning jobs find in the cloud a natural deployment platform to satisfy their notoriously large resource requirements. Yet, to achieve cost efficiency, it is crucial to identify a deployment configuration that satisfies user-defined QoS constraints (e.g., on execution time), while avoiding unnecessary over-provisioning. This paper introduces Lynceus, a new approach for the optimization of cloud based data analytic jobs that improves overstate-of-the-art approaches by enabling significant cost savings both in terms of the final recommended configuration and of the optimization process used to recommend configurations. Unlike existing solutions, Lynceus optimizes in a joint fashion both the cloud-related and the application-level parameters. This allows for a reduction of the cost of recommended configurations by up to 3.7x at the 90-th percentile with respect to existing approaches, which treat the optimization of cloud-related and application-level parameters as two independent problems. Further, Lynceus reduces the cost of the optimization process (i.e., the cloud cost incurred for testing configurations) by up to 11x. Such an improvement is achieved thanks to two mechanisms: i) a timeout approach which allows to abort the exploration of configurations that are deemed suboptimal, while still extracting useful information to guide future explorations and to improve its predictive model – differently from recent works, which either incur the full cost for testing suboptimal configurations or are unable to extract any knowledge from aborted runs; ii) a long-sighted and budget-aware technique that determines which configurations to test by predicting the long-term impact of each exploration – unlike state-of-the-art approaches for the optimization of cloud jobs, which adopt greedy optimization methods.