Juan L. Aparicio Ojea

Juan L. Aparicio Ojea

Berkeley, California, United States
14K followers 500+ connections

Join to follow

About

Founder and CEO of Reshape Automation. On a journey to deploy 10x more robotics and…

Contributions

Activity

Join now to see all activity

Experience

  • Reshape Automation Graphic

    Reshape Automation

    San Francisco Bay Area

  • -

    San Francisco Bay Area

  • -

    San Francisco Bay Area

  • -

    Berkeley

  • -

    Princeton, New Jersey

  • -

    Princeton (NJ)

Education

Volunteer Experience

  • Association for Advancing Automation Graphic

    Board Member - AI Tech Strategy Board

    Association for Advancing Automation

    - Present 6 years 2 months

    The Association for Advancing Automation is THE global advocate for the benefits of automating. We promote automation technologies and ideas that transform the way business is done.

  • Berkeley SkyDeck Graphic

    Advisor

    Berkeley SkyDeck

    - Present 5 years 10 months

    SkyDeck combines the consulting know-how of traditional accelerators with the vast resources of its research university.
    This robust partnership is coupled with SkyDeck’s unique mentorship program to create a powerful environment for startups. Its SkyAdvisors, Partners, and large network of accredited investors connect their SkyTeams to the expertise and capital they need to launch and grow…to the moon.

  • ARM - Advanced Robotics for Manufacturing Graphic

    Member of the Technical Advisory Committee

    ARM - Advanced Robotics for Manufacturing

    - Present 6 years 9 months

    Science and Technology

    Working together with the top robotic minds in the U.S. to create a roadmap of innovation that accelerated the advancement of transformative robotic technologies and education to increase U.S. global manufacturing competitiveness.

  • The Open Group Graphic

    Open Process Automation Forum (OPAF) - Project Manager

    The Open Group

    - 2 years 1 month

    Science and Technology

    The Open Process Automation™ Forum is focused on developing a standards-based, open, secure, interoperable process control architecture.

  • Chair

    SAGA - Siemens Corp. Asian Employee Resource Group

    - 4 years 4 months

    SAGA is a professional organization that brings together members of the Asian community and heritage, as well as employees with an affinity for those cultures; engaging them in activities related to cultural awareness, networking and corporate citizenship.

    Example of involvement:

    • Led China/Germany business practices workshop to help support cross-border collaboration.
    • Pioneered assessment for SAGA executive sponsors by integrating organization’s goals with business…

    SAGA is a professional organization that brings together members of the Asian community and heritage, as well as employees with an affinity for those cultures; engaging them in activities related to cultural awareness, networking and corporate citizenship.

    Example of involvement:

    • Led China/Germany business practices workshop to help support cross-border collaboration.
    • Pioneered assessment for SAGA executive sponsors by integrating organization’s goals with business targets
    • Expanded the organization from a single campus to a multicampus group; and linked it to other Asian organizations within Siemens.

  • Genewin Graphic

    Collaborator

    Genewin

    - Present 4 years 9 months

    Science and Technology

    He colaborado en el área de trabajo de Transición Energética y Reindustrialización del País de Genewin (www.genewin.es) junto a +100 directivos y expertos. El trabajo colaborativo ha tenido como entregable un informe de 5 Ideas para España para impulsar la digitalización de la salud, que puedes descargar aquí: https://genewin.es/transicion-energetica/ - Genewin, sociedad civil al servicio del País, es un proyecto social para impulsar la colaboración de las mentes brillantes de nuestro país, con…

    He colaborado en el área de trabajo de Transición Energética y Reindustrialización del País de Genewin (www.genewin.es) junto a +100 directivos y expertos. El trabajo colaborativo ha tenido como entregable un informe de 5 Ideas para España para impulsar la digitalización de la salud, que puedes descargar aquí: https://genewin.es/transicion-energetica/ - Genewin, sociedad civil al servicio del País, es un proyecto social para impulsar la colaboración de las mentes brillantes de nuestro país, con el objetivo de proponer ideas innovadoras en los 6 principales focos estratégicos, encuadrados en los pilares de Next Generation de la Unión Europea. Los impulsores de la iniciativa son: Endeavor, Fundación Transforma España, South Summit, Lanzadera, Telefónica Open Future, TechCity Barcelona y Opinno

Publications

  • Learning Robotic Assembly from CAD.

    ICRA 2018

    In this work, motivated by recent manufacturing trends, we investigate autonomous robotic assembly. Industrial assembly tasks require contact-rich manipulation skills, which are challenging to acquire using classical control and motion planning approaches. Consequently, robot controllers for assembly domains are presently engineered to solve a particular task, and cannot easily handle variations in the product or environment. Reinforcement learning (RL) is a promising approach for autonomously…

    In this work, motivated by recent manufacturing trends, we investigate autonomous robotic assembly. Industrial assembly tasks require contact-rich manipulation skills, which are challenging to acquire using classical control and motion planning approaches. Consequently, robot controllers for assembly domains are presently engineered to solve a particular task, and cannot easily handle variations in the product or environment. Reinforcement learning (RL) is a promising approach for autonomously acquiring robot skills that involve contact-rich dynamics. However, RL relies on random exploration for learning a control policy, which requires many robot executions, and often gets trapped in locally suboptimal solutions. Instead, we posit that prior knowledge, when available, can improve RL performance. We exploit the fact that in modern assembly domains, geometric information about the task is readily available via the CAD design files. We propose to leverage this prior knowledge by guiding RL along a geometric motion plan, calculated using the CAD data. We show that our approach effectively improves over traditional control approaches for tracking the motion plan, and can solve assembly tasks that require high precision, even without accurate state estimation. In addition, we propose a neural network architecture that can learn to track the motion plan, and generalize the assembly controller to changes in the object positions.

    See publication

  • Dex-Net 2.0: Deep Learning to Plan Robust Grasps with Synthetic Point Clouds and Analytic Grasp Metrics

    Robotics: Science and Systems 2017

    To reduce data collection time for deep learning of robust robotic grasp plans, we explore training from a synthetic dataset of 6.7 million point clouds, grasps, and robust analytic grasp metrics generated from thousands of 3D models from Dex-Net 1.0 in randomized poses on a table. We use the resulting dataset, Dex-Net 2.0, to train a Grasp Quality Convolutional Neural Network (GQ-CNN) model that rapidly predicts the probability of success of grasps from depth images, where grasps are specified…

    To reduce data collection time for deep learning of robust robotic grasp plans, we explore training from a synthetic dataset of 6.7 million point clouds, grasps, and robust analytic grasp metrics generated from thousands of 3D models from Dex-Net 1.0 in randomized poses on a table. We use the resulting dataset, Dex-Net 2.0, to train a Grasp Quality Convolutional Neural Network (GQ-CNN) model that rapidly predicts the probability of success of grasps from depth images, where grasps are specified as the planar position, angle, and depth of a gripper relative to an RGB-D sensor. Experiments with over 1,000 trials on an ABB YuMi comparing grasp planning methods on singulated objects suggest that a GQ-CNN trained with only synthetic data from Dex-Net 2.0 can be used to plan grasps in 0.8s with a success rate of 93% on eight known objects with adversarial geometry and is 3x faster than registering point clouds to a precomputed dataset of objects and indexing grasps. The Dex-Net 2.0 grasp planner is also the highest performing method on a dataset of 10 novel rigid objects and achieves 99% precision (one false positive out of 69 grasps classified as robust) on a dataset of 40 novel household objects, some of which are articulated or deformable.

    See publication

  • Design of parallel-jaw gripper tip surfaces for robust grasping

    ICRA 2017

    Parallel-jaw robot grippers can grasp almost any object and are ubiquitous in industry. Although the shape, texture, and compliance of gripper jaw surfaces affect grasp robustness, almost all commercially available grippers provide a pair of rectangular, planar, rigid jaw surfaces. Practitioners often modify these surfaces with a variety of ad-hoc methods such as adding rubber caps and/or wrapping with textured tape. This paper explores data-driven optimization of gripper jaw surfaces over a…

    Parallel-jaw robot grippers can grasp almost any object and are ubiquitous in industry. Although the shape, texture, and compliance of gripper jaw surfaces affect grasp robustness, almost all commercially available grippers provide a pair of rectangular, planar, rigid jaw surfaces. Practitioners often modify these surfaces with a variety of ad-hoc methods such as adding rubber caps and/or wrapping with textured tape. This paper explores data-driven optimization of gripper jaw surfaces over a design space based on shape, texture, and compliance using rapid prototyping. In total, 37 jaw surface design variations were created using 3D printed casting molds and silicon rubber. The designs were evaluated with 1377 physical grasp experiments using a 4-axis robot (with automated reset). These tests evaluate grasp robustness as the probability that the jaws will acquire, lift, and hold a training set of objects at nominal grasp configurations computed by Dex-Net 1.0. Hill-climbing in parameter space yielded a grid pattern of 0.03 inch void depth and 0.0375 inch void width on a silicone polymer with durometer of A30. We then evaluated performance of this design using an ABB YuMi robot grasping a set of eight difficult-to-grasp 3D printed objects in 80 grasps with four gripper surfaces. The factory-provided gripper tips succeeded in 28.7% of the 80 trials, increasing to 68.7% when the tips were wrapped with tape. Gripper tips with gecko-inspired surfaces succeeded in 80.0% of trials, and gripper tips with the designed silicone surfaces succeeded in 93.7% of trials.

    See publication

  • Exploiting Road Traffic Data for Very Short Term Load Forecasting in Smart Grids

    IEEE Innovative Smart Grid Applications 2014

    If accurate short term prediction of electricity consumption is available, the Smart Grid infrastructure can rapidly and reliably react to changing conditions. The economic importance of accurate predictions justifies research for more complex forecasting algorithms. This paper proposes road traffic data as a new input dimension that can help improve very short term load forecasting. We explore the dependencies between power demand and road traffic data and evaluate the predictive power of the…

    If accurate short term prediction of electricity consumption is available, the Smart Grid infrastructure can rapidly and reliably react to changing conditions. The economic importance of accurate predictions justifies research for more complex forecasting algorithms. This paper proposes road traffic data as a new input dimension that can help improve very short term load forecasting. We explore the dependencies between power demand and road traffic data and evaluate the predictive power of the added dimension compared with other common features, such as historical load and temperature profiles.

    Other authors

  • ParkZoom: A Parking Spot Identification System

    IEEE Intelligent Transportation Systems Conference 2013

    Accurate localization in outdoor and indoor spaces is a challenging task. The widely used GPS is not designed for high accuracy applications and yields accuracy levels not sufficient for lane or spot level localization. In addition, errors from inertial sensors accumulate with time due to integration drift. We introduce a smartphone based, infrastructure aided parking localization system called ParkZoom for estimating (zooming into) the precise parking spot location of a vehicle during…

    Accurate localization in outdoor and indoor spaces is a challenging task. The widely used GPS is not designed for high accuracy applications and yields accuracy levels not sufficient for lane or spot level localization. In addition, errors from inertial sensors accumulate with time due to integration drift. We introduce a smartphone based, infrastructure aided parking localization system called ParkZoom for estimating (zooming into) the precise parking spot location of a vehicle during traversal in both indoor and outdoor parking lots. On the vehicle side, the proposed method utilizes conventional smartphones for generating and transferring continuous sensor data, such as accelerometer, gyroscope, and compass readings. On the infrastructure side, ParkZoom employs statistical learning of sensor data signatures, pattern classification of data, constraint propagation and error correction for accurate parking spot identification. The paper presents experimental results with the ParkZoom algorithm obtained on real data in city driving and two parking areas.

    Other authors

  • Intersection Information Services: A Flexible and Extensible Connected Vehicle Framework

    IEEE International Conference on Connected Vehicles & Expo – ICCVE 2012

    The Connected Vehicle initiative is aiming to revolutionize our way of driving and how we presently solve transportation problems. Its ultimate goal is to create a smarter, safer, and cleaner transportation network, with technologies that will allow vehicles (manually or autonomously) to be interconnected with their surroundings. Through collaborations with car manufacturers and emergency vehicle technology providers in recent years, Siemens has demonstrated the value of the transportation…

    The Connected Vehicle initiative is aiming to revolutionize our way of driving and how we presently solve transportation problems. Its ultimate goal is to create a smarter, safer, and cleaner transportation network, with technologies that will allow vehicles (manually or autonomously) to be interconnected with their surroundings. Through collaborations with car manufacturers and emergency vehicle technology providers in recent years, Siemens has demonstrated the value of the transportation connectivity in car efficiency and safety applications. In this paper, we present our Connected Vehicle platform in order to bring forward information for companies, agencies, and organizations willing to take their next steps into the Connected Vehicle realm.

    Other authors

  • iPASS: Intelligent Pavement Signaling System

    IEEE Intelligent Transportation Systems Conference

    Traditional radio frequency based wireless communication/localization suffers from environmental effects such as multipath propagation. We introduce a robust wireless communication methodology which utilizes the physical world itself to convey information. Our proposed intelligent pavement signaling system (iPASS) can intelligently place irregularities on pavement in order to encode different messages that can be deciphered in real time by an application running on e.g., a phone. iPASS utilizes…

    Traditional radio frequency based wireless communication/localization suffers from environmental effects such as multipath propagation. We introduce a robust wireless communication methodology which utilizes the physical world itself to convey information. Our proposed intelligent pavement signaling system (iPASS) can intelligently place irregularities on pavement in order to encode different messages that can be deciphered in real time by an application running on e.g., a phone. iPASS utilizes available accelerometers, e.g., in today's smartphones, that can detect even small vibrations from the inside of a car that are produced by driving over irregularities on the road. Our feasibility study illustrates that iPASS can successfully decode a 5 bit message using a conventional smartphone when cruising at various speeds.

    Other authors

  • Assessing communications technology options for smart grid applications

    Smart Grid Communications (SmartGridComm), 2011 IEEE International Conference on

    Utilities are at a crossroads in addressing challenges to architect communication networks that can support a robust blend of smart grid applications while simultaneously meeting stringent financial and regulatory objectives. This paper discusses challenges in understanding the communication network choices when supporting applications such as Distribution Automation, Advanced Metering, Automated Demand Response and Electric Vehicle Charging. We also introduce the unique capabilities of a tool…

    Utilities are at a crossroads in addressing challenges to architect communication networks that can support a robust blend of smart grid applications while simultaneously meeting stringent financial and regulatory objectives. This paper discusses challenges in understanding the communication network choices when supporting applications such as Distribution Automation, Advanced Metering, Automated Demand Response and Electric Vehicle Charging. We also introduce the unique capabilities of a tool designed to evaluate different communications technology choices under specific application, network, topology, and geographical constraints.

    Other authors
    • Nazif Tas
    • Amar Patel
    • Michael Loiacono
    • Justinian Rosca

  • Scalable Smart Grid Simulations in OPNET Modeler

    OPNETWORK 2011

    The Smart Grid (SG) communication network is expected to be robust enough to handle various applications with unique requirements in terms of frequency of packets, packet size, delay, etc. Large scale simulations of SG communications, if feasible, would be valuable to identify the robustness issues. We have developed a toolkit that leverages OPNET Modeler’s environment to evaluate a mix of Smart Grid applications under specific network, topology and geographical constraints. In this paper we…

    The Smart Grid (SG) communication network is expected to be robust enough to handle various applications with unique requirements in terms of frequency of packets, packet size, delay, etc. Large scale simulations of SG communications, if feasible, would be valuable to identify the robustness issues. We have developed a toolkit that leverages OPNET Modeler’s environment to evaluate a mix of Smart Grid applications under specific network, topology and geographical constraints. In this paper we discuss scale-up concerns when approaching large scale simulations, which capture environment characteristics such us terrain profile and population density, to mimic realistic results.

    Other authors
    • Amar Patel
    • Nazif Tas
    • Michael Loiacono
    • Justinian Rosca

  • Simulation modelling and analysis of a realistic radio channel model for V2V communications

    Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT), 2010 International Congress on

    Realistic radio channel models are crucial for the success of vehicle-to-vehicle (V2V) system investigations. In this paper, we describe the important parameters for V2V channel modelling and summarize relevant measurement campaigns. Also, we develop and evaluate an ns-3 simulation model of a realistic V2V channel model that is based on measurements in Sweden. The channel model incorporates the effects of obstacles between the transmitter and the receiver. We show that the resulting path loss…

    Realistic radio channel models are crucial for the success of vehicle-to-vehicle (V2V) system investigations. In this paper, we describe the important parameters for V2V channel modelling and summarize relevant measurement campaigns. Also, we develop and evaluate an ns-3 simulation model of a realistic V2V channel model that is based on measurements in Sweden. The channel model incorporates the effects of obstacles between the transmitter and the receiver. We show that the resulting path loss is very different compared with available models

    Other authors

  • Combining power and communication network simulation for cost-effective smart grid analysis

    IEEE COMMUN. SURVEYS & TUTORIALS – SPECIAL ISSUE ON ENERGY AND SMART GRID

    Today’s electricity grid is transitioning to a so-called smart grid. The associated challenges and funding initiatives have spurred great efforts from the research community to propose innovative smart grid solutions. To assess the performance of possible solutions, simulation tools offer a cost effective and safe approach. In this paper we will provide a comprehensive overview of various tools and their characteristics, applicable in smart grid research: we will cover both the communication…

    Today’s electricity grid is transitioning to a so-called smart grid. The associated challenges and funding initiatives have spurred great efforts from the research community to propose innovative smart grid solutions. To assess the performance of possible solutions, simulation tools offer a cost effective and safe approach. In this paper we will provide a comprehensive overview of various tools and their characteristics, applicable in smart grid research: we will cover both the communication and associated ICT infrastructure, on top of the power grid. First, we discuss the motivation for the development of smart grid simulators, as well as their associated research questions and design challenges. Next, we discuss three types of simulators in the smart grid area: power system simulators, communication network simulators, and combined power and communication simulators. To summarize the findings from this survey, we classify the different simulators according to targeted use cases, simulation model level of detail, and architecture. To conclude, we discuss the use of standards and multi-agent based modeling in smart grid simulation.

    Other authors

Patents

  • Network Element Consolidation for Rapid Discrete Event Network Simulations

    US 61/527,207

    Other inventors
    • Amar Patel
    • Michael Loiacono
    • Nazif Tas
    • Justinian Rosca

  • Network Traffic Profile Aggregation for Efficient Discrete Event Smart Grid Network Simulations

    US 61/527,212

    Other inventors
    • Michael Loiacono
    • Amar Patel
    • Nazif Tas
    • Justinian Rosca

  • Smart Grid Communication Assessment and Co-Simulation Tool

    US 61/527,211

    Other inventors
    • Michael Loiacono
    • Amar Patel
    • Nazif Tas
    • Justinian Rosca

Languages

  • Spanish

    Native or bilingual proficiency

  • English

    Full professional proficiency

  • Swedish

    Limited working proficiency

  • Italian

    Professional working proficiency

  • French

    Elementary proficiency

Recommendations received

2 people have recommended Juan L.

Join now to view

More activity by Juan L.

View Juan L.’s full profile

  • See who you know in common
  • Get introduced
  • Contact Juan L. directly
Join to view full profile

Other similar profiles

Explore more posts

Explore collaborative articles

We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.

Explore More

Add new skills with these courses

See all courses