Visual Servoing of Unknown Objects for Family Service Robots

Aiming at satisfying the increasing demand of family service robots for housework, this paper proposes a robot visual servoing scheme based on the randomized trees to complete the visual servoing task of unknown objects in natural scenes. Here, “unknown” means that there is no prior information on object models, such as template or database of the object. Firstly, an object to be manipulated is randomly selected by user prior to the visual servoing task execution. Then, the raw image information about the object can be obtained and used to train a randomized tree classifier online. Secondly, the current image features can be computed using the well-trained classifier. Finally, the visual controller can be designed according to the error of image feature, which is defined as the difference between the desired image features and current image features. Five visual positioning of unknown objects experiments, including 2D rigid object and 3D non-rigid object, are conducted on a MOTOMAN-SV3X six degree-of-freedom (DOF) manipulator robot. Experimental results show that the proposed scheme can effectively position an unknown object in complex natural scenes, such as occlusion and illumination changes. Furthermore, the developed robot visual servoing scheme has an excellent positioning accuracy within 0.05 mm positioning error.

Reinforcement learning for robotic manipulation using simulated locomotion demonstrations

Mastering robotic manipulation skills through reinforcement learning (RL) typically requires the design of shaped reward functions. Recent developments in this area have demonstrated that using sparse rewards, i.e. rewarding the agent only when the task has been successfully completed, can lead to better policies. However, state-action space exploration is more difficult in this case. Recent RL approaches to learning with sparse rewards have leveraged high-quality human demonstrations for the task, but these can be costly, time consuming or even impossible to obtain. In this paper, we propose a novel and effective approach that does not require human demonstrations. We observe that every robotic manipulation task could be seen as involving a locomotion task from the perspective of the object being manipulated, i.e. the object could learn how to reach a target state on its own. In order to exploit this idea, we introduce a framework whereby an object locomotion policy is initially obtained using a realistic physics simulator. This policy is then used to generate auxiliary rewards, called simulated locomotion demonstration rewards (SLDRs), which enable us to learn the robot manipulation policy. The proposed approach has been evaluated on 13 tasks of increasing complexity, and can achieve higher success rate and faster learning rates compared to alternative algorithms. SLDRs are especially beneficial for tasks like multi-object stacking and non-rigid object manipulation.

Tactics for Forsaken Territories: Infrastructural Catalysis Within Disturbed Residual Spaces

<p><b>The perceived dichotomy between the industrial and the ecological or amenity has led to a loss or misperception of identity and value of industrial landscapes. Conventional industrial precinct greening moves or the design of parks within these spaces fail to establish any sense of identity or contribute to the growth and development of these environments. Looking beyond the notion of a park as a respite from the urban condition, the challenge lies in developing parks that capitalise on what is perceived as negative and exploit it as amenity.</b></p> <p>In search of a strategy, the discourse on the relationship between landscape and infrastructure and the ensuing paradigm shift in the way we understand infrastructure, is examined. What has conventionally existed as a mono-programmatic object for the sake of managing a technical problem is being redefined into a multi-layered spatial field, performing over time as well as space. However care must be taken in how we go about redefining the notion of infrastructure; when we keep broadening its definition it begins to lose significance. If infrastructure is to be reinterpreted from a rigid object to a field that is able to engage with open and unpredictable systems, rather than defining what an infrastructural thing may be, it becomes more important to define how something might perform or develop ‘infrastructurally’. Much of the discourse surrounding landscape infrastructure focuses on the efficiency that is to be gained by the layering of multiple flexible systems or employing it as a means to remediate a site. While many contemporary landscape infrastructure projects seek to reintroduce the ecological histories that have been suppressed by urban development, their attempts to do so often erase much of what is too readily dismissed as negative, and with it meaningful social histories and qualities that may be exploited as opportunity.</p> <p>In focussing on the disturbed and residual spaces and the opportunities these territories offer, this thesis seeks to explore the potential of designing infrastructurally to not only reintegrate these forgotten spaces in the urban fabric of their industrial context but to coordinate their development and/or their deterioration in such a way that they become fundamental to the area’s identity and growth.</p> <p>Drawing on Stan Allen’s propositions for infrastructure and reinterpreting them through the lens of landscape as a catalytic infrastructure, an architectural strategy is proposed that capitalises upon the qualities found within the abandoned landscapes of the Seaview/Gracefield industrial precinct in Wellington, New Zealand, and recognises them as an opportunity to develop the concept of park in this context into something that reflects the important social histories of these sites while also presenting a proving ground for future operations. These spaces aim to question the way in which we assess infrastructural efficiency, their performance valued not just in quantitative output but also in qualitative terms.</p>

Tactics for Forsaken Territories: Infrastructural Catalysis Within Disturbed Residual Spaces

<p><b>The perceived dichotomy between the industrial and the ecological or amenity has led to a loss or misperception of identity and value of industrial landscapes. Conventional industrial precinct greening moves or the design of parks within these spaces fail to establish any sense of identity or contribute to the growth and development of these environments. Looking beyond the notion of a park as a respite from the urban condition, the challenge lies in developing parks that capitalise on what is perceived as negative and exploit it as amenity.</b></p> <p>In search of a strategy, the discourse on the relationship between landscape and infrastructure and the ensuing paradigm shift in the way we understand infrastructure, is examined. What has conventionally existed as a mono-programmatic object for the sake of managing a technical problem is being redefined into a multi-layered spatial field, performing over time as well as space. However care must be taken in how we go about redefining the notion of infrastructure; when we keep broadening its definition it begins to lose significance. If infrastructure is to be reinterpreted from a rigid object to a field that is able to engage with open and unpredictable systems, rather than defining what an infrastructural thing may be, it becomes more important to define how something might perform or develop ‘infrastructurally’. Much of the discourse surrounding landscape infrastructure focuses on the efficiency that is to be gained by the layering of multiple flexible systems or employing it as a means to remediate a site. While many contemporary landscape infrastructure projects seek to reintroduce the ecological histories that have been suppressed by urban development, their attempts to do so often erase much of what is too readily dismissed as negative, and with it meaningful social histories and qualities that may be exploited as opportunity.</p> <p>In focussing on the disturbed and residual spaces and the opportunities these territories offer, this thesis seeks to explore the potential of designing infrastructurally to not only reintegrate these forgotten spaces in the urban fabric of their industrial context but to coordinate their development and/or their deterioration in such a way that they become fundamental to the area’s identity and growth.</p> <p>Drawing on Stan Allen’s propositions for infrastructure and reinterpreting them through the lens of landscape as a catalytic infrastructure, an architectural strategy is proposed that capitalises upon the qualities found within the abandoned landscapes of the Seaview/Gracefield industrial precinct in Wellington, New Zealand, and recognises them as an opportunity to develop the concept of park in this context into something that reflects the important social histories of these sites while also presenting a proving ground for future operations. These spaces aim to question the way in which we assess infrastructural efficiency, their performance valued not just in quantitative output but also in qualitative terms.</p>

Impact Device for Biomechanics of Human Head-Neck Injuries

This paper describes experimental tests in LARM2 in Rome to analyze impacts on a human head. The tests consist of performing three different types of impact by hitting a commercial head mannequin with a rigid object. Inertial Measurement Unit (IMU) sensors and force sensors measure each impact’s effect and evaluate the results. The sensors are located on suitable head points to monitor force, acceleration, and angular displacement on small and large lateral impact and top impact events. Results of tests are discussed to investigate and characterize the biomechanics in human head impacts. Considerations from results are used to formulate a new criterion for head-neck injury by impacts.

A fast region-based active contour for non-rigid object tracking and its shape retrieval

Conventional tracking approaches track objects using a rectangle bounding box. Gait, gesture and many medical analyses require non-rigid shape extraction. A non-rigid object tracking is more difficult because it needs more accurate object shape and background separation in contrast to rigid bounding boxes. Active contour plays a vital role in the retrieval of image shape. However, the large computation time involved in contour tracing makes its use challenging in video processing. This paper proposes a new formation of the region-based active contour model (ACM) using a mean-shift tracker for video object tracking and its shape retrieval. The removal of re-initialization and fast deformation of the contour is proposed to retrieve the shape of the desired object. A contour model is further modified using a mean-shift tracker to track and retrieve shape simultaneously. The experimental results and their comparative analysis concludes that the proposed contour-based tracking succeed to track and retrieve the shape of the object with 71.86% accuracy. The contour-based mean-shift tracker resolves the scale-orientation selection problem in non-rigid object tracking, and resolves the weakness of the erroneous localization of the object in the frame by the tracker.