Real-time perception meets reactive motion generation

被引：41

作者：

Kappler D. ^{[1
,2
]}

Meier F. ^{[1
,3
,4
]}

Issac J. ^{[1
,2
]}

Mainprice J. ^{[1
,5
]}

Cifuentes C.G. ^{[1
]}

Wuthrich M. ^{[1
]}

Berenz V. ^{[1
]}

Schaal S. ^{[1
,3
]}

Ratliff N. ^{[2
]}

Bohg J. ^{[1
,6
]}

机构：

[1] Autonomous Motion Department, MPI for Intelligent Systems, Stuttgart

[2] Lula Robotics Inc., Seattle, 98102, WA

[3] CLMC Lab, University of Southern California, Los Angeles, 90007, CA

[4] Department of Computer Science and Engineering, University of Washington, Seattle, 98195, WA

[5] University of Stuttgart, Stuttgart

[6] Department of Computer Science, Stanford University, Stanford, 94305, CA

来源：

IEEE Robotics and Automation Letters | 2018年 / 3卷 / 03期

关键词：

perception for grasping and manipulation; Reactive and sensor-based planning; sensor-based control;

D O I：

10.1109/LRA.2018.2795645

中图分类号：

学科分类号：

摘要：

We address the challenging problem of robotic grasping and manipulation in the presence of uncertainty. This uncertainty is due to noisy sensing, inaccurate models, and hard-to-predict environment dynamics. We quantify the importance of continuous, real-time perception and its tight integration with reactive motion generation methods in dynamic manipulation scenarios. We compare three different systems that are instantiations of the most common architectures in the field: 1) a traditional sense-plan-act approach that is still widely used; 2) a myopic controller that only reacts to local environment dynamics; and 3) a reactive planner that integrates feedback control and motion optimization. All architectures rely on the same components for real-time perception and reactive motion generation to allow a quantitative evaluation. We extensively evaluate the systems on a real robotic platform in four scenarios that exhibit either a challenging workspace geometry or a dynamic environment. We quantify the robustness and accuracy that is due to integrating real-time feedback at different time scales in a reactive motion generation system. We also report on the lessons learned for system building. © 2016 IEEE.

引用

页码：1864 / 1871

页数：7

共 36 条

[1]

Leonard J., Et al., A perception-driven autonomous urban vehicle, J. Field Robot., 25, 10, pp. 727-774, (2008)

[2]

Richter C., Bry A., Roy N., Polynomial Trajectory Planning for Aggressive Quadrotor Flight in Dense Indoor Environments, pp. 649-666, (2016)

[3]

Atkeson C., Et al., No falls, no resets: Reliable humanoid behavior in the DARPA robotics challenge, Proc. 2015 IEEE-RAS. 15th Int. Conf. Humanoid Robots, pp. 623-630, (2015)

[4]

Johnson M., Et al., Team ihmc's lessons learned from the DARPA robotics challenge trials, J. Field Robot., 32, 2, pp. 192-208, (2015)

[5]

Eppner C., Et al., Lessons from the amazon picking challenge: Four aspects of robotic systems building, Proc. Robot., Sci. Syst., pp. 4831-4835, (2016)

[6]

Correll N., Et al., Analysis and observations from the first amazon picking challenge, IEEE Trans. Autom. Sci. Eng., 15, 1, pp. 172-188, (2018)

[7]

Wuthrich M., Pastor P., Kalakrishnan M., Bohg J., Schaal S., Probabilistic object tracking using a range camera, Proc. 2013 IEEE/RSJ Int. Conf. Intell. Robots Syst., pp. 3195-3202, (2013)

[8]

Garcia Cifuentes C., Issac J., Wuthrich M., Schaal S., Bohg J., Probabilistic articulated real-time tracking for robot manipulation, IEEE Robot. Autom. Lett., 2, 2, pp. 577-584, (2017)

[9]

Ratliff N., Toussaint M., Schaal S., Understanding the geometry of workspace obstacles in motion optimization, Proc. 2015 IEEE Int. Conf. Robot. Autom., pp. 4202-4209, (2015)

[10]

Mainprice J., Ratliff N., Schaal S., Warping the workspace geometry with electric potentials for motion optimization of manipulation tasks, Proc. 2016 IEEE/RSJ Int. Conf. Intell. Robots Syst., pp. 3156-3163, (2016)

← 1 2 3 4 →