Team IHMC's Lessons Learned from the DARPA Robotics Challenge Trials
Team IHMC's Lessons Learned from the DARPA Robotics Challenge Trials Abstract: This article is a summary of the experiences of the Florida Institute for Human & Machine Cognition (IHMC) team during the DARPA Robotics Challenge (DRC) Trials. The primary goal of the DRC is to develop robots capable of assisting humans in responding to natural and manmade disasters. The robots are expected to use standard tools and equipment to accomplish the mission. The DRC Trials consisted of eight different challenges that tested robot mobility, manipulation, and control under degraded communications and time constraints. Team IHMC competed using the Atlas humanoid robot made by Boston Dynamics. We competed against 16 international teams and placed second in the competition. This article discusses the challenges we faced in transitioning from simulation to hardware. It also discusses the lessons learned both during the competition and in the months of preparation leading up to it. The lessons address the value of reliable hardware and solid software practices. They also cover effective approaches to bipedal walking and designing for human-robot teamwork. Lastly, the lessons present a philosophical discussion about choices related to designing robotic systems.
Feb. 2015
Matthew Johnson, Brandon Shrewsbury, Sylvain Bertrand, Tingfan Wu, Daniel Duran, Marshall Floyd, Peter Abeles, Douglas Stephen, Nathan Mertins, Alex Lesman, John Carff, William Rifenburgh, Pushyami Kaveti, Wessel Straatman, Jesper Smith, Maarten Griffioen, Brooke Layton, Tomas de Boer, Twan Koolen, Peter Neuhaus, and Jerry Pratt
Florida Institute for Human & Machine Cognition
Human-in-the-loop Control of a Humanoid Robot for Disaster Response: A Report from the DARPA Robotics Challenge Trials
Human-in-the-loop Control of a Humanoid Robot for Disaster Response: A Report from the DARPA Robotics Challenge Trials Abstract: The DARPA Robotics Challenge (DRC) requires teams to integrate mobility, manipulation, and perception to accomplish several disaster-response tasks. We describe our hardware choices and software architecture, which enable human-in-the-loop control of a 28 degree-of-freedom atlas humanoid robot over a limited bandwidth link. We discuss our methods, results, and lessons learned for the DRC Trials tasks. The effectiveness of our system architecture was demonstrated as the WPI-CMU DRC Team scored 11 out of a possible 32 points, ranked seventh (out of 16) at the DRC Trials, and was selected as a finalist for the DRC Finals.
Feb. 2015
Mathew DeDonato, Velin Dimitrov, Ruixiang Du, Ryan Giovacchini, Kevin Knoedler, Xianchao Long, Felipe Polido, Michael A. Gennert, and Taskin Padir
Worcester Polytechnic Institute
Siyuan Feng, Hirotaka Moriguchi, Eric Whitman, X. Xinjilefu, and Christopher G. Atkeson
The Robotics Institute
Inside the Virtual Robotics Challenge: Simulating Real-Time Robotic Disaster Response
Abstract: This paper presents the software framework established to facilitate cloud-hosted robot simulation. The framework addresses the challenges associated with conducting a task-oriented and real-time robot competition, the Defense Advanced Research Projects Agency (DARPA) Virtual Robotics Challenge (VRC), designed to mimic reality. The core of the framework is the Gazebo simulator, a platform to simulate robots, objects, and environments, as well as the enhancements made for the VRC to maintain a high fidelity simulation using a high degree of freedom and multisensor robot. The other major component used is the CloudSim tool, designed to enhance the automation of robotics simulation using existing cloud technologies. The results from the VRC and a discussion are also detailed in this work.
Jan. 2015
Carlos E. Agüero, Nate Koenig, Ian Chen, Hugo Boyer, Steven Peters, John Hsu, Brian Gerkey, Steffi Paepcke, Jose L. Rivero, Justin Manzo, Eric Krotkov, and Gill Pratt
Open Source Robotics Foundation
Using planar features for fast localization in indoor environments
Abstract: A means for fast and accurate indoor localization using point-cloud data is presented. A sparse environment model is proposed, comprised of a list of planar patches. It is shown that, with this simplified model, incoming point-cloud data can be associated rapidly with one (or none) of the constituent planes, and the environment model can be fit to the clustered sample points algebraically in a least-squares sense in real time.
Oct. 2014
Li Ma, Cheung, E.C.H.,Newman, W.
The University of Hong Kong
Initial Pose Estimation Using Cross-Section Contours
The University of Hong Kong Abstract: This paper presents a means to approximate an object's pose, suitable for initialization of the Iterative Closest Point algorithm. The class of problems considered is objects lying stably on a planar surface, for which a relatively small number of pose types are possible. Within each pose type, the registration problem is reduced to 3 dimensions. Using contours computed from horizontal slices, it is shown that relatively noisy point-cloud samples can yield good estimates of pose. Experimental results using an Atlas robot are presented. The proposed method offers an efficient means to initialize point-cloud fitting, resulting in faster, more reliable convergence.
Oct. 2014
Cheung, E.C.H., Cao Chao, Newman, W.S.
The University of Hong Kong
Continuous Trajectory Estimation for 3D SLAM from Actuated Lidar
Abstract: We extend the Iterative Closest Point (ICP) algorithm to obtain a method for continuous-time trajectory estimation (CICP) suitable for SLAM from actuated lidar. Traditional solutions to SLAM from actuated lidar rely heavily on the accuracy of an auxiliary pose sensor to form rigid frames. These frames are then used with ICP to obtain accurate pose estimates. However, since lidar records a single range sample at time, any error in inter-sample sensor motion must be accounted for. This is not possible if the frame is treated as a rigid point cloud. In this work, instead of ICP we estimate a continuous-time trajectory that takes into account intersample pose errors. The trajectory is represented as a linear combination of basis functions and formulated as a solution to a (sparse) linear system without restrictive assumptions on sensor motion. We evaluate the algorithm on synthetic and real data and show improved accuracy in open-loop SLAM in comparison to state-of-the-art rigid registration methods.
May 2014
Hatem Alismail, L. Douglas Baker and Brett Browning
The Robotics Institute
A Modular Approach for Remote Operation of Humanoid Robots in Search and Rescue Scenarios
Abstract: In this work we present a modular, robust, and user-friendly Pilot Interface meant to control humanoid robots in rescue scenarios during dangerous missions. YARP is used to communicate to low-level hardware components and to interconnect control modules (receive the status and request actions). ROS is used to retrieve many sensors data and to display the robot status. The operator is immersed into a 3D reconstruction of the environment and can manipulate 3D virtual objects. The operator can control the robot at three different levels. The high-level control deals with human-like actions which involve the whole robot's actuation and perception. The mid-level control generates tasks in cartesian space w.r.t. a reference frame on the robot. Finally the low level control operates in joint space.
May 2014
Alessandro Settimi, Valerio Varricchio, Enrico Mingo Hoffman, Alessio Rocchi, Nikos G. Tsagarakis, Antonio Bicchi
Istituto Italiano di Tecnologia, Italy
Corrado Pavan, Mirko Ferrati, Kamilo Melo
Universita di Pisa, Italy
Dual Arm Impedance Control with a Compliant Humanoid: Application to a Valve Turning Task
Abstract: In this work we present a modular, robust, and user-friendly Pilot Interface meant to control humanoid robots in rescue scenarios during dangerous missions. YARP is used to communicate to low-level hardware components and to interconnect control modules (receive the status and request actions). ROS is used to retrieve many sensors data and to display the robot status. The operator is immersed into a 3D reconstruction of the environment and can manipulate 3D virtual objects. The operator can control the robot at three different levels. The high-level control deals with human-like actions which involve the whole robot's actuation and perception. The mid-level control generates tasks in cartesian space w.r.t. a reference frame on the robot. Finally the low level control operates in joint space.
May 2014
 Arash Ajoudani, Jinoh Lee, Alessio Rocchi, Mirko Ferrati, Enrico Mingo Hoffman, Aelssandro Settimi, Nikos G. Tsagarakis, Darwin. G. Caldwell, and Antonio Bicchi
Learning Algorithms and Systems Laboratory
Plane Detection and Segmentation For DARPA Robotics Challange
Abstract: The purpose of this project is to perform a comparative study of plane detection algorithms by altering the computational process flow and filter parameters to determine the impact on the results and performance. We evaluate several different processing workflows by testing each preprocessing chain with varying parameters for each component in the chain. The end goal is to be able to find the best combination of process chain and filter parameters to get the best result as quickly as possible. The implementation relies on the Point Cloud Library (PCL) in Robot Operating System (ROS). The testing has been performed by collecting sensor data from WPI’s DARPA Robotics Challenge hardware platform, Atlas.
Apr. 2014
Jacob H. Palnick
Worcester Polytechnic Institute
Competing in the DARPA Virtual Robotics Challenge as the SARBOT Team
Abstract: On October 25th, 2012 DARPA anounced the DARPA Robotics Challenge (DRC) kick off. The DRC is part of the US Department of Defense’s strategic plan to conduct humanitarian, disaster relief and related operations. The DRC particularly aims to promoting innovation in robotic technology (software and hardware) to improve the capability of ground robotic systems for disaster-response operations. Designed as a three-stage competition, the DRC started with a Virtual Robotics Challenge (VRC), where a selection of 26 teams proved their algorithms in a cloud-based simulated platform. The six best-performing teams in the VRC competition won an ATLAS humanoid robot to continue to the second stage of the DRC. This paper describes the work performed by one of the VRC competitors, the SARBOT Team.
2014
Elena Garcia, Manuel Ferre, Juan C. Arevalo, Daniel Sanz-Merodio
Centre for Automation and Robotics, Spain
Manuel Ocaña, Luis Miguel Bergasa, Eduardo J. Molinos, Noelia Hernandez, Ángel Llamazares
Universidad de Alcala, Spain
Francisco Suarez
Universidad Politechnica de Madrid, Spain
Mohamed Abderrahim, Silvia Rodriguez
Universidad Carlos III de Madrid, Spain
Analysis of Flat Terrain for the Atlas Robot
Automatic Calibration of a Range Sensor and Camera System Abstract: This paper gives a description of an approach to analyze the sensor information of the surroundings to select places where the foot of a humanoid can be placed. This will allow apply such robot in a rescue scenario, as foreseen in the DARPA Robotics Challenge, where a robot is forced to traverse difficult terrain. I
Oct. 2012
Maarten de Waard, Maarten Inja, and Arnoud Visser
Universiteit van Amsterdam
Automatic Calibration of a Range Sensor and Camera System
Automatic Calibration of a Range Sensor and Camera System Abstract: We propose an automated method to recover the full calibration parameters between a 3D range sensor and a monocular camera system. Our method is not only accurate and fully automated, but also relies on a simple calibration target consisting of a single circle. This allows the algorithm to be suitable for applications requiring in-situ calibration. We demonstrate the effectiveness of the algorithm on a cameralidar system and show results on 3D mapping tasks.
Oct. 2012
Hatem Alismail, L. Douglas Baker, Brett Browning
National Robotics Engineering Center
A Tele-Operation Tool for Humanoid Robots: On the Pilot Interface Design and Functionality
Alessandro Settimi, Corrado Pavan, Mirko Ferrati Antonio Bicchi
Universita di Pisa, Italy
Enrico Mingo Hoffman, Alessio Rocchi, Nikos G. Tsagarakis
Istituto Italiano di Tecnologia, Italy
Shared Autonomy: An Operator’s Perspective
Abstract: This paper analyzes the human-robot boundaries chosen by team TRACLabs at the DRC Trials from the perspective of the human operator and presents ideas for future improvements based on our experience at the Trials.
Joshua James
TRACLabs
Tight Coupling between Manipulation and Perception using SLAM
Abstract:A tight coupling between perception and manipulation is required for dynamic robots to react in a timely and appropriate manner to changes in the world. In conventional robotics, perception transforms visual information into internal models which are used by planning algorithms to generate trajectories for motion. Under this paradigm, it is possible for a plan to become stale if the robot or environment changes configuration before the robot can replan. Perception and actuation are only loosely coupled through planning; there is no rapid feedback or interplay between them. For a statically stable robot in a slowly changing environment, this is an appropriate strategy for manipulating the world. A tightly coupled system, by contrast, connects perception directly to actuation, allowing for rapid feedback. This tight coupling is important for a dynamically unstable robot which engages in active manipulation. In such robots, planning does not fall between perception and manipulation; rather planning creates the connection between perception and manipulation. We show that Simultaneous Localization and Mapping (SLAM) can be used as a tool to perform the tight coupling for a humanoid robot with numerous proprioceptive and exteroceptive sensors. Three different approaches to generate a motion plan for grabbing a piece of debris is evaluated using for Atlas humanoid robot. Results indicate higher success rate and accuracy for motion plans that implement tight coupling between perception and manipulation using SLAM.
Benzun Pious Wisely Babu, Christopher Bove, Michael A. Gennert
Worcester Polytechnic Institute
Human-Supervised Control of the ATLAS Humanoid Robot for Traversing Doors
Abstract:Door traversal is generally a trivial task for human beings but particularly challenging for humanoid robots. This paper describes a holistic approach for a full-sized humanoid robot to traverse through a door in an outdoor unstructured environment as specified by the requirements of the DARPA Robotics Challenge. Door traversal can be broken down into four sub-tasks; door detection, walk to the door, door opening, and walk through the door. These topics are covered in detail along with the approaches used for step planning and motion planning. The approach presented in this paper can be further extended to a wide range of door types and configurations.
Nandan Banerjee, Xianchao Long, Ruixiang Du, Michael Gennert, and Taskin Padir
Worcester Polytechnic Institute
Felipe Polido, Siyuan Feng, Christopher G. Atkeson
The Robotics Institute