史上最全 2019 ICRA顶会四足机器人文献整理
一、ICRA论文集中相关文献对应subsession时间
15:15-16:30, Subsession TuCT1-14, 220
Legged Robots I - 2.3.14 Interactive
Session, 6 papers
15:15-16:30, Subsession TuCT1-25, 220
Legged Robots II - 2.3.25 Interactive
Session, 6 papers
09:40-10:55, Subsession WeAT1-14, 220
Legged Robots III - 3.1.14 Interactive
Session, 6 papers
11:30-12:45, Subsession WeBT1-14, 220
Legged Robots IV - 3.2.14 Interactive
Session, 6 papers
16:00-17:15, Subsession WeCT1-14, 220
Legged Robots V - 3.3.14 Interactive Session,
6 papers
二、文献整理内容
Legged Robots I
1.1 SpaceBok: A
Dynamic Legged Robot for Space Exploration（ETH Zurich）
Philip Arm, Radek Zenkl, Patrick
Barton, Lars Beglinger, Alex Dietsche,
Luca Ferrazzini, Elias Hampp, Jan
Hinder, Camille Huber, David Schaufelberger,
Felix Schmitt, Benjamin Sun, Boris
Stolz, Hendrik Kolvenbach and Marco Hutter
本文介绍了一种用于探测低重力天体的四足机器人spacebok。机器人的腿配置是基于一个优化的平行运动机制，允许整合平行弹性元件来储存和释放能量，以进行强有力的跳跃动作。高扭矩无刷电机与定制的单级行星齿轮变速器结合，可根据电机电流在各触点上进行强制控制。实验证明直接跳(pronk)被证明是低重力运动的有效解决方案。
Robotic Systems Lab | ETH Zurich（https://rsl.ethz.ch/）：
[1]Marko Bjelonic, Lorenz Wellhausen,
Marco Hutter and Navinda Kottege，Walking Posture Adaptation for Legged Robot Navigation in
Confined Spaces，Russell Buchanan, Tirthankar
Bandyopadhyay, IEEE Robotics and Automation Letters, vol. 4: no. 2, pp.
2148-2155, Piscataway, NJ: IEEE, 2019.
[2]Hendrik Kolvenbach, Christian Bärtschi,
Lorenz Wellhausen, Ruben Grandia and Marco Hutter， Haptic Inspection of Planetary Soils with Legged Robots，IEEE Robotics and Automation Letters, vol. 4:
no. 2, pp. 1626-1632, New York, NY: IEEE, 2019
[3]Jemin Hwangbo, Joonho Lee, Alexey
Dosovitskiy, Carmine Dario Bellicoso, Vassilios Tsounis, Vladlen Koltun and
Marco Hutter，Learning agile and dynamic motor skills
for legged robots，Science Robotics,
vol. 4: no. 26, pp. eaau5872, Washington, DC: American Association for the
Advancement of Science (AAAS), 2019.
在模拟中训练神经网络策略并将其传输到最先进的腿系统的方法，从而利用快速、自动化和经济高效的数据生成方案。该方法被应用于Anymal机器人，一个复杂的中型狗大小的四足系统。使用模拟训练的策略，四足机器人实现了超越以往方法的运动技能：Anymal能够精确高效地执行高水平的身体速度指令，比以前更快地运行，并从摔倒中恢复。
[4]Jan Carius, René Ranftl, Vladlen Koltun
and Marco Hutter，Trajectory Optimization for Legged Robots
with Slipping Motions，IEEE Robotics and
Automation Letters, Piscataway, NJ: IEEE, 2019.
[5]C. Dario Bellicoso, Marko Bjelonic,
Lorenz Wellhausen, Kai Holtmann, Fabian Günther, Marco Tranzatto, Peter
Fankhauser and Marco Hutter，Advances in
real-world applications for legged robots，Journal of Field Robotics, vol. 35: no. 8, pp. 1311-1326,
Hoboken, NJ: John Wiley & Sons, 2018
[6] C. Dario Bellicoso, Fabian Jenelten,
Christian Gehring and Marco Hutter，Dynamic locomotion through online nonlinear motion
optimization for quadrupedal robots，IEEE Robotics and Automation Letters, vol. 3: no. 3, pp.
2261-2268, Piscataway, NJ: IEEE, 2018.
[7]Michael Neunert, Markus Stäuble, Markus
Giftthaler, Carmine Dario Bellicoso, Jan Carius, Christian Gehring, Marco
Hutter and Jonas Buchli，Whole-Body
Nonlinear Model Predictive Control Through Contacts for Quadrupeds，IEEE Robotics and Automation Letters, vol. 3:
no. 3, pp. 1458-1465, New York, NY: IEEE, 2018.
[8]Marco Hutter, Christian Gehring,
Andreas Lauber, Fabian Gunther, Carmine D. Bellicoso, Vassilios Tsounis, Péter
Fankhauser, Remo Diethelm, Samuel Bachmann, Michael Blösch, Hendrik Kolvenbach,
Marko Bjelonic, Linus Isler and Konrad Meyer，ANYmal - toward legged robots for harsh environments，Advanced Robotics, vol. 31: no. 17, pp.
918-931, London: Taylor & Francis, 2017
[9] Jemin Hwangbo, Inkyu Sa, Roland
Siegwart and Marco Hutter，Control of a
quadrotor with reinforcement learning，IEEE Robotics and Automation Letters, vol. 2: no. 4, pp.
2096-2103, New York, NY: IEEE, 2017.
[10] Hendrik Kolvenbach, Giorgio
Valsecchi, Ruben Grandia, Antoni Ruiz, Fabian Jenelten and Marco Hutter，Tactile Inspection of Concrete Deterioration
in Sewers with Legged Robots，12th Conference on
Field and Service Robotics (FSR 2019), Tokyo, Japan, August 29-31, 2019.
[11] Vivian Medeiros S., Marko Bjelonic,
Edo Jelavic, Roland Siegwart, Marco Meggiolaro A. and Marco Hutter，Trajectory Optimization for Wheeled
Quadrupedal Robots Driving in Challenging Terrain，9th International Symposium on Adaptive Motion of Animals
and Machines (AMAM 2019), Lausanne, Switzerland, August 20-23, 2019.
[12] Timon Homberger, Lorenz Wellhausen,
Péter Fankhauser and Marco Hutter，Support Surface Estimation for Legged Robots，IEEE International Conference on Robotics and
Automation (ICRA), Montreal, Canada Zurich: ETH Zurich.
[13] Hendrik Kolvenbach, Dario Bellicoso,
Fabian Jenelten, Lorenz Wellhausen and Marco Hutter，Efficient Gait Selection for Quadrupedal
Robots on the Moon and Mars，14th International
Symposium on Artificial Intelligence, Robotics and Automation in Space
(i-SAIRAS 2018), Madrid, Spain Noordwijk: ESA Conference Bureau, June 4-6,
2018.
[14] Hendrik Kolvenbach, Manuel
Breitenstein, Christian Gehring and Marco Hutter，Scalability Analysis of Legged Robots for Space
Exploration，68th International Astronautical Congress
(IAC 2017), Adelaide, Australia, pp.10399-10413, Red Hook, NY: Curran,
September 25-29, 2017.
[15] Thomas Bi, Péter Fankhauser, Dario
Bellicoso and Marco Hutter，Real-Time Dance
Generation to Music for a Legged Robot，25th IEEE/RSJ International Conference on Intelligent
Robots and Systems (IROS), Madrid, Spain, pp.1038-1044, Piscataway, NJ: IEEE,
October 1-5, 2018.
[16]Marko Bjelonic, C. Dario Bellicoso, M.
Efe Tiryaki and Marco Hutter，Skating with a
Force Controlled Quadrupedal Robot，IEEE/RSJ International Conference on Intelligent Robots
and Systems (IROS 2018), Madrid, Spain, pp.7555-7561, Piscataway, NJ: IEEE,
October 1-5, 2018.
[17] Péter Fankhauser, Marko Bjelonic,
Dario Bellicoso, Takahiro Miki and Marco Hutter，Robust Rough-Terrain Locomotion with a Quadrupedal Robot，IEEE International Conference on Robotics and
Automation (ICRA 2018), Brisbane, Australia, pp.5761-5768, Piscataway, NJ:
IEEE, May 21-25, 2018.[18] Towards a Passive Adaptive Planar Foot with Ground
Orientation and Contact Force Sensing for Legged RobotsRoman Käslin, Hendrik
Kolvenbach, Laura Paez, Klajd Lika and Marco Hutter25th IEEE/RSJ International
Conference on Intelligent Robots and Systems (IROS), Madrid, Spain,
pp.2707-2714, Piscataway, NJ: IEEE, October 1-5, 2018.
[19] Foot-Eye Calibration of Legged Robot
KinematicsFabian Blöchliger, Michael Blösch, Péter Fankhauser, Marco Hutter and
Roland Siegwart19th International Conference on Climbing and Walking Robots
(CLAWAR 2016), London, England, pp.420-427, Singapore: World Scientific,
September 12-14, 2016.
1.2 Mini Cheetah:
A Platform for Pushing the Limits of Dynamic Quadruped Control(MIT Biomimetic
Robotics Lab)
Benjamin Katz1, Jared Di Carlo2 and Sangbae Kim1
一种新型的轻量化、低成本的四足机器人” 迷你猎豹”, 四足机器人的第一次基于离线轨迹优化的后空翻（backflips generated
by offline trajectory optimization,）用凸模型预测控制（CMPC）演示了机器人的动态快跑、快跑、跳跃和内旋步态，速度可达每秒2.45米。
MIT Biomimetic Robotics Lab（http://biomimetics.mit.edu/）:
[1] Boussema, C., M. J. Powell, G. Bledt,
A. J. Ijspeert, P. M. Wensing, and S. Kim, “Gait Emergence and Disturbance
Recovery for Legged Robots via the Feasible Impulse Set”, IEEE Robotics
and Automation Letters, 2019 ,submitted
[2] Wensing, P. M., A. Wang, S. Seok, D.
Otten, J. Lang, and S. Kim, “Proprioceptive actuator design in the MIT
cheetah: Impact mitigation and high-bandwidth physical interaction for dynamic
legged robots”, IEEE Transactions on Robotics, 2017.
麻省理工学院猎豹腿的弹簧处理冲击，使得它能够在动态跳跃过程中控制接触力，接触时间低至85 ms，峰值力超过450 N。猎豹的独特能力，在高速三维跑动和跳跃中实现强大的冲击力控制操作。
[3]Kim, S., and P. M. Wensing,
“Design of Dynamic Legged Robots”, Foundations and Trends® in
Robotics, vol. 5, pp. 117–190, 2017.
在陆地、海洋和空气中动物所表现出的非凡的运动能力的启发下，机器人工程师几十年来一直在努力在腿部机器中实现类似的动态运动能力。本文也利用了仿生学的一些思想，着重介绍了腿型机器人的机械设计，旨在为今后新机构的发展以及上述耦合问题提供参考。
[4]Park, H.-W., Wensing, P. M., & Kim,S. (2017). High-speed bounding with the MIT Cheetah 2: Control design and experiments. The International Journal of Robotics Research, 36(2), 167–192. https://doi.org/10.1177/0278364917694244
本文介绍了麻省理工学院猎豹2的边界控制器(bounding controller)的设计与实现及其实验结果。控制框架在硬件中提供稳定的边界，速度高达6.4 m/s，总运输成本最低为0.47。这些结果是无约束实验四足机在效率和速度方面前所未有的成就。
[5] Hyun, D. J., Lee, J., Park, S., & Kim, S. (2016). Implementation of trot-to-gallop transition and subsequent gallop on the MIT Cheetah I. The International Journal of Robotics Research,
35(13), 1627–1650. https://doi.org/10.1177/0278364916640102
本文展示了机器人四足动物从小跑到疾驰的过渡和随后的稳定疾驰。麻省理工学院猎豹I是一个平面四足平台，用于高速运行，在跑步机上以3.2 m/s（弗劳德数为2.1）的速度完成这些任务。相位图表明，所提出的控制器在快跑和疾驰两种情况下都能实现稳定的极限环，从而使快跑能够在高速状态下转换为疾驰步态。
[6] Bledt, G., M. J. Powell, B. Katz, J. Di Carlo, P. M. Wensing, and S. Kim, “MIT Cheetah 3: Design and Control of a Robust, Dynamic Quadruped Robot”, IEEE/RSJ International Conference on Intelligent Robotics (IROS), 10/2018.
猎豹3号可以单独使用接触检测来运行、攀爬甚至跳跃。仅仅依靠触碰算法来“感觉”它的周围。
[7] Ramos, J., and S. Kim, “Dynamic Bilateral Teleoperation of the Cart-Pole: A Study Toward the Synchronization of Human Operator and Legged Robot”, IEEE/RSJ International Conference on Intelligent Robotics (IROS), 10/2018.
[8] Di Carlo, J., P… M. Wensing, B. Katz, G. Bledt, and S. Kim, “Dynamic Locomotion in the MIT Cheetah 3 Through Convex Model-Predictive Control”, IEEE/RSJ International Conference on Intelligent Robotics (IROS), 10/2018.
[9] Bledt, G., P. M. Wensing, S. Ingersoll, and S. Kim, “Contact Model Fusion for Event-Based Locomotion in Unstructured Terrains”, 2018 IEEE International Conference on Robotics and
Automation (ICRA) Finalist Best Overall Paper & Finalist Best Student Paper, May, 2018.
[10] Ramos, J., and S. Kim, “Facilitating Model-Based Control through Software-Hardware
Co-Design”, IEEE International Conference on Robotics and Automation (ICRA), 05/2018.
[11]Bledt, G., P. M. Wensing, and S. Kim, “Policy-regularized model predictive control to stabilize diverse quadrupedal gaits for the MIT cheetah”, 2017 IEEE/RSJ International
Conference on Intelligent Robots and Systems (IROS), Sept, 2017.
[12]. Ramos, J., and S. Kim, “Improving humanoid posture Teleoperation by Dynamic Synchronization through operator motion anticipation”, IEEE International Conference on
Robotics and Automation (ICRA), 06/2017.
1.4 《Stanford Doggo：一种开源的，准直接驱动的四足机器人》 Nathan Kau, Aaron Schultz, Natalie Ferrante, Patrick Slade
文章介绍了Doggo——一种四足机器人的准直接驱动设计方法。它具有很强的动态运动能力，在垂直跳跃灵活性方面，其垂直运动速度比之前最先进的腿部机器人有明显提高。
[1] Andrew D. Wilson, Jarvis A. Schultz, Alex R. Ansari & Todd D. Murphey. Dynamic Task
Execution using Active Parameter Identification with the Baxter Research Robot [J]. IEEE Transactions on Automation Science and Engineering, 2017, 14(1), 391-397.
文章给出了基于Baxter研究机器人的系统模型实时参数估计和动态任务轨迹优化的实验结果。利用闭环非线性控制技术，实时地利用最大Fisher信息的主动估计器进行实时控制。
[2] Anastasia Mavrommati, Jarvis A. Schultz & Todd D. Murphey. Real-time Dynamic-Mode
Scheduling Using Single-Integration Hybrid Optimization for Linear Time-Varying Systems [J], IEEE Transactions on Automation Science and Engineering, 2016, 13(3), 1385-1398.
文章研究线性时变切换系统在二次成本函数作用下的实时模式调度问题。提出了一种基于投影的方法，使得在开环优化过程中不需要进行仿真，且优于数值积分方法。之后进行了一些应用。
[3] Patrick Slade, Siobhan Powell & Michael F. Howland. Optimal control of a single leg hopper
by Liouvillian system reduction, 2017.
文章对单腿跳跃机器人在整个跳跃运动中采用最优控制的效果进行了评价。在不考虑物理约束的情况下，根据Raibert控制框架，通过手调PD控制器实现基本控制。
1.5 《带身体姿势控制的工作空间中央处理器组(CPG)，可在全方位移动过程中实现稳定的定向视觉》 Samuel Shaw, Guillaume Sartoretti, Jake Olkin, William Paivine, Howie Choset
文中研究了视觉系统安装在高自由度腿部机器人的身体上（且视觉系统和机器人身体之间没有任何附加的自由度）以进行主动感知。并展示了如何利用机器人的附加DOF来实现独立的运动和视线控制。
[1] Guillaume Sartoretti, Samuel Shaw, Katie Lam, Naixin Fan, Matthew Travers & Howie
Choset. Central Pattern Generator With Inertial Feedback for Stable Locomotion and Climbing in Unstructured Terrain [C]. 2018 IEEE International Conference on Robotics and Automation (ICRA), 2018.
中央模式发生器（CPG）模型可用于设计关节机器人的步态，在这些模型中加入步态适应的感觉反馈可以提高机器人在复杂地形中的运动性能。文章提出了一种新的将惯性反馈纳入CPG框架的方法，用于在陡峭的非结构化地形上控制腿运动时的身体姿势。
[2] Francesco Ruscelli, Guillaume Sartoretti, Junyu Nan, Zhixin Feng, Matthew Travers & Howie Choset. Proprioceptive-Inertial Autonomous Locomotion for Articulated Robots [C]. 2018 IEEE nternational Conference on Robotics and Automation (ICRA), 2018.
文中提出了一种基于力和惯性信息的自主运动模块化框架来快速应对碰撞障碍物。由一种基于双稳态动力学系统的新型并联控制器，能连续调节机器人的运动方向。
[3] Guillaume Sartoretti, Yunfei Shi, William Paivine, Matthew Travers & Howie Choset. Distributed Learning for the Decentralized Control of Articulated Mobile Robots [C]. 2018 IEEE International Conference on Robotics and Automation (ICRA), 2018.
分散控制体系结构能够独立地协调关节体的空间分布，以实现系统级目标。文中提出了一种算法结构来提供在单个平台上实现分布式学习，从而得到高效的硬件实现。
[4] Guillaume Sartoretti, Justin Kerr, Yunfei Shi, Glenn Wagner, T. K. Satish Kumar, Sven Koenig & Howie Choset. PRIMAL: Pathfinding via Reinforcement and Imitation Multi-Agent Learning [J]. IEEE Robotics and Automation Letters, 2018, 4(3), 2378 - 2385.
文章提出了PRIMAL——一种多代理寻径(Multi-agent path finding)框架，它扩展了分布式学习的工作，可以更好地规划在线路径。
1.6 《频率感知模型预测控制》 Ruben Grandia, Farbod Farshidian, Alexey Dosovitskiy, René Ranftl, Marco Hutter
当轨迹优化充分利用所提供的模型来执行动态任务时，未建模动态的存在使运动在实际的机器人系统上不可行。文中提出了频率型成本函数，以在腿部机器人最优控制的背景下实现鲁棒解。使机器人在规划的运动、扭矩和力轨迹上的跟踪性能得到了显著改善。
[1] Ruben Grandia, Diego Pardo & Jonas Buchli. Contact Invariant Model Learning for Legged Robot Locomotion [J]. IEEE Robotics and Automation Letters, 2018, 3(3), 2291 - 2298.
文章提出了一个新的研究腿部机器人执行运动任务的动力学公式。建立了不依赖于接触力子空间中的运动方程，从而可以在不需要力传感器数据的情况下建立学习问题。
[2] Ruben Grandia, Farbod Farshidian, René Ranftl & Marco Hutter. Feedback MPC for
Torque-Controlled Legged Robots [C]. 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2019.
文章证明了基于差分动态规划(DDP)的模型预测控制(MPC)算法的反馈策略是一个可行的替代方案，可以弥补低MPC更新率和驱动指令率之间的差距。
[3] Farbod Farshidian, Edo Jelavic, Asutosh Satapathy, Markus Giftthaler & Jonas Buchli. Real-time motion planning of legged robots: A model predictive control approach [C]. 2017 IEEE-RAS 17th International Conference on Humanoid Robotics, 2017.
文章介绍了一种实时、约束、非线性模型预测控制(MPC)在腿部机器人运动规划中的应用。该方法采用了一种被称为SLQ的约束最优控制算法，通过引入一种多处理方案来估计后向传递中的值函数，提高了算法的效率。
[4] Farbod Farshidian, Edo Jelavić, Alexander W. Winkler & Jonas Buchli. Robust whole-body motion control of legged robots [C]. 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2017.
文章介绍了一种鲁棒控制的手臂和腿力矩控制机器人的全身运动控制结构。该方法基于接触力的鲁棒控制，以跟踪规划的质心轨迹。尽管有刚体模型不匹配、延迟、接触面刚度和未观察到的地面轮廓，但它仍能保证稳健的稳定性和性能。
[5] Farbod Farshidian, Michael Neunert, Alexander W. Winkler, Gonzalo Rey & Jonas Buchli. An efficient optimal planning and control framework for quadrupedal locomotion [C]. 2017 IEEE International Conference on Robotics and Automation, 2017.
提出了一种有效的动态规划框架，用于腿部机器人的优化规划和控制。首先将此问题表示为切换系统的最优控制问题，提出了一种多层次的优化方法，以确定最优切换时间和最优连续控制输入。
[6] Alexander W. Winkler, Farbod Farshidian, Michael Neunert, Diego Pardo & Jonas Buchli. Online walking motion and foothold optimization for quadruped locomotion [C]. 2017 IEEE International Conference on Robotics and Automation, 2017.
文章提出了一种不使用显式足迹规划器的四足机器人步行运动生成算法，该算法通过同时优化质心轨迹和落脚点。通过在与零力矩点相关的质心上施加稳定性约束，并在立足点和质心位置之间明确地实施运动学约束来实现。
[7] Michael Neunert, Farbod Farshidian, Alexander W. Winkler & Jonas Buchli. Trajectory Optimization Through Contacts and Automatic Gait Discovery for Quadrupeds [J]. IEEE Robotics and Automation Letters, 2017, 2(3), 1502 - 1509.
在这封信中，我们提出了一个通过接触进行全身运动规划的轨迹优化框架。我们演示了该方法如何在四足机器人上自动发现不同的步态和动态运动。与大多数以前的方法相比，我们不预先指定接触开关、计时、点或步态模式，但它们是优化的直接结果。
[8] Russell Buchanan, Tirthankar Bandyopadhyay, Marko Bjelonic, Lorenz Wellhausen, Marco Hutter & Navinda Kottege. Walking Posture Adaptation for Legged Robot Navigation in onfined Spaces [J]. IEEE Robotics and Automation Letters, 2019, 4(2), 2148 - 2155.
文章提出了一个可伴随着地图和规划策略使腿部机器人能够自主改变其身体形状，以在有限的空间导航的方法。利用机器人携带的距离传感器生成以机器人为中心的多高程地图，实现了对地形的测绘。
[9] Roman Käslin, Hendrik Kolvenbach, Laura Paez, Klajd Lika & Marco Hutter. Towards a Passive Adaptive Planar Foot with Ground Orientation and Contact Force Sensing for Legged Robots [C]. 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2018.
适应地面可以利用地形的结构，使腿的运动保持稳定的立足点。因此，文中提出了一种具有三个旋转自由度的被动自适应平面脚部，这种脚很轻，因此适合高度动态的腿部机器人。
[10] Peter Fankhauser, Marko Bjelonic, C. Dario Bellicoso, Takahiro Miki & Marco Hutter. Robust Rough-Terrain Locomotion with a Quadrupedal Robot [C]. 2018 IEEE International Conference on Robotics and Automation, 2018.
文章提出了一种适用于四足机器人感知粗糙地形运动的运动规划方法。规划者通过利用获取的地形图找到安全的立足点以及无碰撞的摆动腿运动。由此提出了一种新的姿态优化方法，使机器人能够爬过主要的障碍物。
[11] C. Dario Bellicoso, Fabian Jenelten, Christian Gehring & Marco Hutter. Dynamic Locomotion Through Online Nonlinear Motion Optimization for Quadrupedal Robots [J]. IEEE Robotics and Automation Letters, 2018, 3(3), 2261 - 2268.
文中提出了一种实时运动规划和控制方法，使四足机器人能够执行包括小跑、步速和动态横向行走在内的动态步态，以及具有跳跃、内旋和跑步小跑等全飞行阶段的步态。该方法还可以使步态之间平稳过渡。
Legged Robots II
2.1 《Mitigating energy loss in a robot hopping on a physically emulated dissipative substrate》Sonia Roberts and Daniel E. Koditschek
本文提出了一个地面模拟器，测试了假设:通过编程虚拟阻尼力来减缓运输的能量成本可以减轻Minitaur脚进入模拟颗粒介质的影响。对基板仿真器进行编程，以展示以线性刚度和二次阻尼为特征的粒状介质简化体积行为模型的机制，与标称控制器相比，实现了20％的一致节能，节省高达50％具体条件。
[1] F. Qian, D. Jerolmack, N. Lancaster, G. Nikolich, P. Reverdy, S. F. Roberts, T. Shipley, R. S. Van Pelt, T. M. Zobeck, and D. E. Koditschek, “Ground robotic measurement of aeolian processes,” Aeolian Research, vol. 27, pp. 1–11, 2017.
[2] G. Kenneally, A. De, and D. E. Koditschek, “Design principles for a family of direct-drive legged robots,” IEEE Robotics and Automation Letters, vol. 1, no. 2, pp. 900–907, 2016.
[3] G. C. Haynes, J. Pusey, R. Knopf, A. M. Johnson, and D. E. Koditschek, “Laboratory on legs: an architecture for adjustable morphology with legged robots,” in Unmanned Systems Technology XIV, vol. 8387, p. 83870W, International Society for Optics and Photonics, 2012.
[4] C. M. Hubicki, J. J. Aguilar, D. I. Goldman, and A. D. Ames, “Tractable terrain-aware motion planning on granular media: an impulsive jumping study,” in Intelligent Robots and Systems (IROS), 2016 IEEE/RSJ International Conference on, pp. 3887–3892, IEEE, 2016.
[5] M. H. Raibert, Legged robots that balance. MIT press, 1986.
[6] G. Kenneally and D. E. Koditschek, “Leg design for energy management in an lectromechanical robot,” in 2015 IEEE/RSJ International Conference on Intelligent Robots and
Systems (IROS), pp. 5712–5718, IEEE, 2015.
2.2 《Energy Efficient Navigation for Running Legged Robots》Mario Y. Harper, John V. Nicholson, Emmanuel G. Collins, Jr.2, Jason Pusey3 and Jonathan E. Clark2
本文提出了基于抽样的模型预测优化(SBMPO)与基于采样的搜索相结合的方法，直接利用动力学模型，适用于基于步态的运动，使其能够在保持自然动力学和自我稳定性的同时生成最佳运动轨迹。
[1] F. Qian, D. Jerolmack, N. Lancaster, G. Nikolich, P. Reverdy, S. F. Roberts, T. Shipley, R.
S. Van Pelt, T. M. Zobeck, and D. E. Koditschek, “Ground robotic measurement of
aeolian processes,”
[2] H.-W. Park, P. M. Wensing, and S. Kim, “High-speed bounding with the mit cheeah 2: Control design and experiments,”
[3] D. J. Blackman, J. V. Nicholson, J. L. Pusey, M. P. Austin, C. Young, J. M. Brown,
and J. E. Clark, “Leg design for running and jumping dynamics,” .
[5] M. Hutter, C. Gehring, A. Lauber, F. Gunther, C. D. Bellicoso, V. Tsounis, P. Fankhauser, R.
Diethelm, S. Bachmann, M. Blösch et al., “Anymal-toward legged robots for harsh environments,”
[6] M. Wermelinger, P. Fankhauser, R. Diethelm, P. Krüsi, R. Siegwart, and M. Hutter, “Navigation planning for legged robots in challenging terrain,”
[7] K. Hauser, T. Bretl, J.-C. Latombe, K. Harada, and B. Wilcox, “Motion planning for legged robots on varied terrain,”
[8]G. C. Haynes, J. Pusey, R. Knopf, A. M. Johnson, and D. E. Koditschek, “Laboratory on legs: an architecture for adjustable morphology with legged robots,”
[9] L. Ding, H. Gao, Z. Deng, J. Song, Y. Liu, G. Liu, and K. Iagnemma, “Footâ˘AS¸terrain interaction mechanics for legged robots: Modeling and experimental validation,”
2.3 《Force-controllable Quadruped Robot System with Capacitive-type Joint Torque Sensor》
Yoon Haeng Lee,
Young Hun Lee, Hyunyong Lee, Hansol Kang, Luong Tin Phan, Sungmoon Jin, Yong
Bum Kim, Dong-Yeop Seok, Seung Yeon Lee, Hyungpil Moon, Ja Choon Koo, and Hyouk
Ryeol Choi
作者及其团队开发出一个受力可控的四足机器人系统，该系统通过使用廉价位置编码器（τ= k·4θ）与致动器串联放置的弹性元件的变形，提供强大而出色的扭矩/力控制，可以抵御外部冲击，而无需使用昂贵的传感器。
[1] K. Loffler, M. Gienger, F. Pfeiffer, “Sensors and control concept of a biped robot."
[2]Y. H. Lee, et al., “Development of Torque Controllable Leg for RunningRobot,AiDIN-IV,”
[3]Hubicki, Christian, et al., “ATRIAS: Design and validation of a tetherfree 3D-capable spring-mass bipedal robot,"
[4]Seok, S., et al., “Design principles for highly efficient quadrupeds and implementation on
the MIT Cheetah robot,”
[5] G. Kenneally, A. De, and D. E. Koditschek, “Design Principles for a Family of Direct-Drive
Legged Robots,”
2.4《Fast and Continuous Foothold Adaptation for Dynamic Locomotionthrough CNNs》
Octavio Villarreal1, Victor Barasuol, Marco Camurri, Luca Franceschi,Michele Focchi,
Massimiliano Pontil, Darwin G. Caldwell and Claudio Semini
本文提出了一种基于视觉反馈的实时动态立足点适应策略。为了对来自环境的视觉刺激作出反应，弥合盲目反应运动和纯粹基于视觉的计划策略之间的差距，运用了基于卷积神经网络（CNN）的自监督立足分类器。并使用了车载计算机和传感器以完全反应的方式调整脚的着陆位置。
[1]Octavio Villarreal, Victor Barasuol, Marco Camurri, Michele Focchi, Luca Franceschi,
Massimiliano Pontil, Darwin G Caldwell, Claudio Semini,”Deep convolutional
terrain assessment for visual reactive footstep correction on dynamic legged
robots ”
[2]Claudio Semini, Victor Barasuol, Thiago Boaventura, Marco Frigerio, Michele Focchi, Darwin G Caldwell, Jonas Buchli,”Towards versatile legged robots through active
impedance control”
本文讨论了通过主动控制关节特别是腿部的机械阻抗而产生的多功能机器人的机会。与诸如弹簧的无源元件相比，通过扭矩控制接头实现有源阻抗，允许实时调节刚度和阻尼。实时可调的刚度和阻尼是实现多功能性的基本构件。并第一次在没有弹簧等被动元件的机器人上成功实施飞行小跑。
[3]C. Semini, V. Barasuol, J. Goldsmith, M. Frigerio, M. Focchi, Gao Y., D. Caldwell,”Design of
the Hydraulically-Actuated, Torque-Controlled Quadruped Robot HyQ2Max”
本文进行了灵活的四足机器人的新颖设计，能够在平坦/不平坦的地形上进行小跑/爬行，平衡和自动扶正; 设计了详细的方法找到合适的液压缸/阀门属性和连杆参数，着重优化执行器区域。
[4] Simona Nobili, Marco Camurri, Victor Barasuol, Michele Focchi, Darwin G. Caldwell, Claudio
Semini, Maurice Fallon,”Heterogeneous Sensor Fusion for Accurate State
Estimation of Dynamic Legged Robots”
本文提出了一个动态步行和小跑四足动物的状态估计系统。该方法融合了四个异构传感器源（惯性，运动学，立体视觉和激光雷达），以在存在诸如滑动和失误等干扰的情况下保持机器人基础链路速度和位置的准确且一致的估计。
[5]Marco Camurri, Maurice Fallon, Stéphane Bazeille, Andreea Radulescu, Victor Barasuol, Darwin
G. Caldwell, Claudio Semini,”Probabilistic Contact Estimation and Impact
Detection for State Estimation of Quadruped Robots”
在本文中提出了一种新的概率接触估计方法，用于无接触传感的动态腿式机器人状态估计。该方法使用逻辑分类器来学习具有最高概率的GRF阈值，以便最小化基本速度误差。
[6]Michele Focchi, Romeo Orsolino, Marco Camurri, Victor Barasuol, Carlos Mastalli, Darwin G
Caldwell, Claudio Semini,”Heuristic Planning for Rough Terrain Locomotion in
Presence of External Disturbances and Variable Perception Quality”
本文提出了一种基于启发式的规划方法，使得四足机器人能够在没有视觉反馈的情况下成功地穿越显着粗糙的地形（例如，直径达10厘米的石头）。当可用时，该方法还允许根据3D地图的质量以多种方式利用视觉反馈（例如，以增强步进策略）。拟议的框架还包括在特定情况下触发的反应，以及在线估计未知的时变扰动并对其进行补偿的可能性。
[7]Elco Heijmink, Andreea Radulescu, Brahayam Ponton, Victor Barasuol, Darwin G. Caldwell,
Claudio Semini,”Learning Optimal Gait Parameters and Impedance Profiles for
Legged Locomotion”
本文提出了一种通过学习一系列参数来改善运动步态表现的方法。专注于通过学习步态参数，阻抗曲线和中继控制器的增益来改善小跑步态的性能。并通过选择提高能源效率（使用CoT测量）的次要目标（在达到期望的行驶速度之后）来解决之前未探索的方面。
[8]Yifu Gao, Victor Barasuol, Darwin G Caldwell, Claudio Semini,”Study on the morphological
parameters of quadruped robot designs considering ditch traversability”
本文研究了四足机器人形态参数与沟渠交叉能力的关系。基于静态稳定性和穿越沟渠时的临界条件，进行了一系列模拟以研究质量和尺寸相关参数的影响。根据所提出的仿真结果，当机器人穿过沟渠时，肢体的尺寸在所有四种膝关节配置的沟渠穿越能力中起主要作用。对于FB和BF膝关节配置，由于结构的对称性，链节的质量变化不影响沟渠交叉的能力;但是对于FF和BB膝盖配置，当链路质量相对于干线增加时，产生的沟渠交叉能力将降低。所以FB和BF是沟渠交叉任务的首选配置。这两种配置的影响在访问和离开步骤中是相反的。躯干的音高将通过改变质量分布和腿部运动范围的方式对所有配置产生影响.
[9]David Wisth, Marco Camurri, Maurice Fallon,”Robust Legged Robot State Estimation Using
Factor Graph Optimization”
本文提出了一种状态估计的因子图优化方法，它紧密地融合和平滑惯性导航，腿部测距和视觉测距，进而减少对滑倒时失败的足部接触分类的依赖性，并减少动态运动（例如小跑）期间的位置漂移。
[10]Andrzej Reinke, Marco Camurri, Claudio Semini,”A Factor Graph Approach to Multi-Camera
Extrinsic Calibration on Legged Robots”
本文提出了一种因子图方法来对四足机器人进行多摄像机校准。 我们已经证明了因子图框架代表了视觉惯性方法的有效且灵活的替代方案，其需要平滑运动和所有轴的平衡激励以提供可靠的结果。 另一方面，视觉运动学方法要求在末端执行器（即，基准标记支撑件）处的硬件修改是精确的
[11]Carlos Mastalli, Michele Focchi, Ioannis Havoutis, Andreea Radulescu, Sylvain Calinon,
Jonas Buchli, Darwin G Caldwell, Claudio Semini,”Trajectory and foothold
optimization using low-dimensional models for rough terrain locomotion”
本文提出了一种直接使用地形信息的粗糙地形上的运动轨迹优化方法。该方法提供最佳CoM运动和相应的最佳立足点位置。此外，该解决方案考虑了动态步行所需的躯干姿态调制。我们采用参数化预览模型，基于随机的探索和后退地平线规划的组合，成功跨越各种具有挑战性的地形。
[12]Bernardo Aceituno-Cabezas, Carlos Mastalli, Hongkai Dai, Michele Focchi, Andreea
Radulescu, Darwin G Caldwell, José Cappelletto, Juan C Grieco, Gerardo
Fernández-López, Claudio Semini,”Simultaneous contact, gait, and motion
planning for robust multilegged locomotion via mixed-integer convex
optimization”
本文提出了一种混合整数凸形公式，以计算有效的方式同时规划接触位置，步态转换和运动。本文方法不仅限于平坦的地形，也不限于预先指定的步态序列。而是结合摩擦锥稳定裕度，近似机器人的扭矩限制，并使用混合整数凸约束计划步态。通过遍历不同的挑战性地形实验验证了对HyQ机器人的方法，其中非凸性和平坦地形假设可能导致次优或不稳定的计划。本文的方法在保持较低计算时间的同时提高了运动鲁棒性。
2.5《Keep Rollin’ – Whole-Body Motion Control and
Planning for Wheeled Quadrupedal Robots》Marko Bjelonic, C. Dario Bellicoso, Yvain de Viragh, Dhionis Sako,F. Dante Tresoldi, Fabian Jenelten and Marco Hutter
本文展示了动态运动策略的四足机器人，它结合了步行和驾驶的优点。开发的优化框架 - 工作紧密地集成了车轮引入的额外自由度。本文方法依赖于基于零点矩的运动优化，其不断更新参考轨迹。参考运动由分层全身控制器跟踪，该控制器通过求解包括非完整滚动约束在内的优先任务的序列来计算最优广义加速度和接触力。
[1]Marco Hutter, Christian Gehring, Andreas Lauber, Fabian Gunther, Carmine Dario Bellicoso,
Vassilios Tsounis, Péter Fankhauser, Remo Diethelm, Samuel Bachmann, Michael
Blösch, Hendrik Kolvenbach, Marko Bjelonic, Linus Isler, Konrad
Meyer,”ANYmal-toward legged robots for harsh environments”
[2]Marko Bjelonic, Navinda Kottege, Philipp Beckerle,”Proprioceptive control of an over-actuated hexapod robot in unstructured terrain”
[3]Timon Homberger, Marko Bjelonic, Navinda Kottege, Paulo VK Borge,”Terrain-dependant
control of hexapod robots using vision”
[4]Marco Hutter, Christian Gehring, Dominic Jud, Andreas Lauber, C Dario Bellicoso, Vassilios
Tsounis, Jemin Hwangbo, Karen Bodie, Peter Fankhauser, Michael Bloesch, Remo
Diethelm, Samuel Bachmann, Amir Melzer, Mark Hoepflinger,”Anymal-a highly
mobile and dynamic quadrupedal robot”
本文设计的机器的设计重点是户外适用性，简单的维护和用户友好的操作，以便在现实世界的场景中实现未来的操作。 使用关节执行器进行的性能测试表明，扭矩控制带宽超过70 Hz，具有高抗干扰能力，以及以最大速度移动时的冲击稳健性。 在一系列实验中证明，ANYmal可以执行步行步态，以中等速度动态地小跑，并且能够进行特殊操作以站立或爬行非常陡峭的楼梯。
[4] Christian Gehring, Stelian Coros, Marco Hutler, Carmine Dario Bellicoso, Huub Heijnen,
Remo Diethelm, Michael Bloesch, Péter Fankhauser, Jemin Hwangbo, Mark
Hoepflinger, Roland Siegwart,”Practice makes perfect: An optimization-based
approach to controlling agile motions for a quadruped robot”
[5] Jemin Hwangbo, Joonho Lee, Alexey Dosovitskiy, Dario Bellicoso, Vassilios Tsounis, Vladlen
Koltun, Marco Hutter,”Learning agile and dynamic motor skills for legged
robots”
[7]C Dario Bellicoso, Koen Kramer, Markus Stäuble, Dhionis Sako, Fabian Jenelten, Fabian
Bjelonic,Marco Hutter,”ALMA-Articulated Locomotion and Manipulation for a
Torque-Controllable Robot”
本文提出了ALMA，一种扭矩控制的四足机器人的运动规划和控制框架，配备六自由度机械臂，能够在执行操作任务时执行动态运动。在线运动规划框架与基于分层优化算法的全身控制器一起使系统能够在执行操作空间末端执行器控制，反应式人机协作和躯干姿势优化的同时行走，小跑和步速，从而增加手臂的工作空间。整个系统的扭矩控制使得能够实现顺应行为，允许用户安全地与机器人交互。
2.6《Computational Design of Robotic Devices
fromHigh-Level Motion Specifications》
Sehoon Ha,Member, IEEE,Stelian Coros,Member, IEEE,Alexander Alspach,Member, IEEE,James M. Bern,Student Member, IEEE,Joohyung Kim,Member, IEEE,Katsu Yamane,Member, IEEE
本文提出了一种新颖的启发式算法，该算法针对用户提供的运动任务优化机器人设备的设计。设计参数包括每肢的链接数，关节类型，链路长度和电机控制信号。关键点是将设计优化表示为一个最短路径发现问题，可以通过由新型启发式函数驱动的A *算法来解决。 通过为各种操纵器和腿式机器人生成优化设计来验证所提出的算法的鲁棒性。 此外，通过与两个基线算法（广度优先搜索和具有简单的基于错误的启发函数的A *算法）进行比较来证明算法的效率。
[1]Sehoon Ha, Stelian Coros, Alexander Alspach, Joohyung Kim, Katsu Yamane,”Joint
Optimization of Robot Design and Motion Parameters using the Implicit Function
Theorem.”
本文提出了一种新的算法，用于优化机器人设计参数，例如链接长度和执行器布局，以及相关的运动参数，包括关节位置，执行器力和每帧的接触力。
[2]Visak CV Kumar, Sehoon Ha, C Karen Liu,”Expanding Motor Skills through Relay Neural Networks”
本文提出了一种技术，可以将复杂的机器人任务分解为更简单的子任务并按顺序训练它们，以便机器人可以逐渐扩展其现有技能。主要思想是建立一个由神经网络代表的本地控制策略树，将其称为中继神经网络。从尝试从一小组初始状态实现任务的根策略开始，每个后续策略通过将新状态驱动到现有“良好”状态来扩展成功初始状态集。本文的算法利用策略的值函数来确定每个策略下的状态是否“良好”。利用许多现有的策略搜索算法，这些算法与策略同时学习价值函数，例如那些使用actor-critic表示或使用优势函数来减少方差的算法。并证明了中继网络可以解决欠驱动动态系统的复杂连续控制问题。
[3] Visak CV Kumar, Sehoon Ha, Katsu Yamane,”Improving model-based balance controllers using
reinforcement learning and adaptive sampling”
本文提出了一个学习框架，通过扩展吸引区域（RoA）来增强基于模型的最优控制器的性能。 训练控制策略时，使用深度强化学习技术在基于模型的控制器之上生成额外的控制信号。
[4]Tuomas Haarnoja, Aurick Zhou, Sehoon Ha, Jie Tan, George Tucker, Sergey Levine,”Learning to walk via deep reinforcement learning”
在本文中，开发了一个稳定的评论深度强化学习算法的变体，它需要最小的超参数调整，同时也只需要适度的试验来学习多层神经网络策略。该算法基于最大熵强化学习的框架，并通过动态地自动调整确定策略随机性的温度参数来自动地对开发进行交易。
[5]Sehoon Ha, Stelian Coros, Alexander Alspach, Joohyung Kim, Katsu Yamane,”Task-based limb optimization for legged robots”
本文提出了这样一个模型来共同设计机器人的运动和腿部配置，从而优化性能的度量。 该框架首先规划了由质心和英尺组成的简化模型的轨迹。 然后，框架优化每个腿部连接的长度，同时解决相关的全身运动。
[6]Sehoon Ha, Stelian Coros, Alexander Alspach, Joohyung Kim, Katsu Yamane,”Computational co-optimization of design parameters and motion trajectories for robotic
systems”
[7]Sehoon Ha, Stelian Coros, Alexander Alspach, James M Bern, Joohyung Kim, Katsu
Yamane,”Computational design of robotic devices from high-level motion
specifications”
[8]Sehoon Ha, Alexander Nicholas Alspach, Joohyung Kim, Katsu Yamane, Stelian
Coros,”Computational design of robots from high-level task specifications”
Legged Robots III
3.1《Realizing Learned Quadruped Locomotion
Behaviors through Kinematic Motion Primitives》Abhik Singla，Shounak Bhattacharya，Dhaivat Dholakiya，Shalabh Bhatnagar，Ashitava Ghosal，Bharadwaj Amrutur，Shishir Kolathaya
作者研究并论证了通过少量基本轨迹的计算足以得出机器人各关节运动的轨迹周期，可以以此来减少机器人关节运动轨迹周期的采样数量，将D-RL从仿真模拟转向实际硬件的过程，主要研究的是数据驱动的KMPS。
[1]Abhik Singla,Partha Pratim Roy,Debi Prosad Dogra,“Visual rendering of shapes on
2D display devices guided by hand gestures”
3.2《Single-shot Foothold Selection and Constraint
Evaluation for Quadruped Locomotion》Dominik Belter, Jakub Bednarek, Hsiu-Chin Lin, Guiyang Xin, Michael Mistry
本文的作者通过运用离线的神经网络学习，让机器人通过估计潜在立足点的几何特性选择最佳的潜在立足点，通过立足点的集合特性以及运动范围进行约束，让机器人具有再山脚环境中选择最佳立足点的学习能力。
[1]Dominik Belter,Jan Wietrzykowski,Piotr Skrzypczyński,“Employing Natural Terrain
Semantics in Motion Planning for a Multi-Legged Robot”
[2]Dominik Belter,Michał Nowicki,“Optimization-based legged odometry and sensor
fusion for legged robot continuous localization”
[3]Dominik Belter,Michał Nowicki,Piotr Skrzypczyński,“Modeling spatial uncertainty of
point features in feature-based RGB-D SLAM”
[4]Michal R Nowicki,Dominik Belter,Aleksander Kostusiak,Petr Cízek,Jan Faigl,Piotr
Skrzypczynski,“An experimental study on feature-based SLAM for
multi-legged robots with RGB-D sensors”
3.3《Optimized Jumping on the MIT Cheetah 3 Robot》Quan Nguyen, Matthew J. Powell, Benjamin Katz,Jared Di Carlo, and Sangbae Kim
本文作者提出利用基于系统的全动力学模型来优化麻省理工学院的猎豹3机器人的跳跃功能，包括高效的轨迹优化、精确的高频跟踪控制器和鲁棒着陆控制器，以稳定机器人碰撞后的位置和方向。
[1]Thomas Albini,Natalia F. Callaway,Glenn J. Jaffe,William Feuer,Janet Davis,Debra
Goldstein,Careen Lowder,Phoebe Lin,Quan Nguyen,Sunil Srivastava,David
Callanan,Anat Galor,Raquel Goldhardt,Harry Flynn,“MUST Beg to Differ”
[2]Toan Dinh,TuanKhoa Nguyen,HoangPhuong Phan,Quan Nguyen,Jisheng Han,Sima
Dimitrijev,NamTrung Nguyen,Dzung Viet Dao,“Thermoresistance of p ‐Type
4H–SiC Integrated MEMS Devices for High‐Temperature SensingThermoresistance of
p ‐Type 4H–SiC Integrated MEMS Devices for High‐Temperature Sensing”
[3]Dirk Tempelaar,Bart Rienties,Quan Nguyen,“A multi-modal study into students’
timing and learning regulation: time is ticking”
[4]Clayton E. Friedman,Quan Nguyen,Samuel W. Lukowski,Abbigail Helfer,Han Sheng Chiu,Jason Miklas,Shiri Levy,Shengbao Suo,Jing-Dong Jackie Han,Pierre Osteil,Guangdun
Peng,Naihe Jing,Greg J. Baillie,Anne Senabouth,Angelika N. Christ,Timothy J.
Bruxner,Charles E. Murry,Emily S. Wong,Jun Ding,Yuliang Wang,James
Hudson,Hannele Ruohola-Baker,Ziv Bar-Joseph,Patrick P.L. Tam,Joseph E.
Powell,Nathan J. Palpant,“Single-Cell Transcriptomic Analysis of Cardiac
Differentiation from Human PSCs Reveals HOPX-Dependent Cardiomyocyte
Maturation”
[5]Maciej Daniszewski,Quan Nguyen,Hun S. Chy,Vikrant Singh,Duncan E. Crombie,Tejal
Kulkarni,Helena H. Liang,Priyadharshini Sivakumaran,Grace E. Lidgerwood,Damian
Hernández,Alison Conquest,Louise A. Rooney,Sophie Chevalier,Stacey B.
Andersen,Anne Senabouth,James C. Vickers,David A. Mackey,Jamie E. Craig,Andrew
L. Laslett,Alex W. Hewitt,Joseph E. Powell,Alice Pébay,“Single cell profiling
identifies key pathways expressed by iPSCs cultured in different commercial
media”
[6]Bart Rienties,Tim Lewis,Ruth McFarlane,Quan Nguyen,Lisette Toetenel,“Analytics
in online and offline language learning environments: the role of learning
design to understand student online engagement”
[7]Dirk Tempelaar,Bart Rienties,Jenna Mittelmeier,Quan Nguyen,“Student profiling
in a dispositional learning analytics application using formative
assessment”
[8]Quan Nguyen,Bart Rienties,Lisette Toetenel,Rebecca Ferguso,Denise
Whitelock,“Examining the designs of computer-based assessment and its
impact on student engagement, satisfaction, and pass rates”
[9]William Rhoades,Drew Dickson,Quan Nguyen,Diana Do,“Management of macular edema due
to central retinal vein occlusion – The role of aflibercept”
[10]Yousef Dhafiri,Khalid Al Rubaie,Omar Kirat,William May,Quan Nguyen,Igor
Kozak,“Multifocal choroiditis with retinal vasculitis, optic neuropathy,
and keratoconus in a young Saudi male”
3.4《CENTAURO: A Hybrid Locomotion and High Power
Resilient Manipulation Platform》Navvab Kashiri,
Lorenzo Baccelliere, Luca Muratore, Arturo Laurenzi, Zeyu Ren,Enrico Mingo
Hoffman, Malgorzata Kamedula, Giuseppe Francesco Rigano,Jorn Malzahn, Stefano
Cordasco, Paolo Guria, Alessio Margan, Nikos G. Tsagaraki
本文作者根据半人马座实验平台设计了一个四足结合混合轮机驱动的高强度密度驱动的一体化实验平台，有效的满足在非结构化的、粗糙环境中的应用需求，是将许多与驱动系统相关的设计和实现原则、车身结构和驱动的集成以及轮盘相结合的结果。
[1]JinohLee,Maolin Jin,Navvab Kashiri,Darwin G. Caldwell,Nikolaos G.
Tsagarakis,“Inversion-free force tracking control of piezoelectric
actuatorsusing fast finite-time integral terminal sliding-mode”
[2]Emmanouiln Spyrakos-Papastavridis,Navvab Kashiri,Peter R.N. Childs,Nikos G.
Tsagarakis,“Online impedance regulation techniques for compliant humanoid
balancing”
Legged Robots IV
4.1 《低摩擦路面下stealth walking步态的产生Generation of Stealth Walking Gait on
Low-Friction Road Surface》 Asano,Fumihiko
作者研究了在没有控制力矩的情况下，利用stealth walking法生成的欠驱动步行者的自适应行走步态。通过应用角动量约束控制（AMCC），实现无摩擦情景（模拟）以及低摩擦情景（现实）下的稳定行走。并通过数学和数值研究，从抗滑特性的角度讨论了双肢支撑阶段上肢控制的优化问题。
[1]F. Asano,“Stealth walking of 3-link planar underactuated biped,” Proc.of the IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, pp.
4118–4124, 2017.
[2]F. Asano, K.Matsuura, S. Kobayashi and Y. Kikuchi, “Motion generation and analysis of high-speed stealth walking on stairs,” Proc. of the 12th IFAC Symp. on Robot Control, 046, 2018.
[3]F.Asano,“Stealth walking with reduced double-limb support phase and its extension to careful legged locomotion,”on Multibody System Dynamics,2018
[4]F. Asano,“High-speed Stealth Walking of Underactuated Biped Utilizing Effects of Upper-body Control and Semicircular Feet,”Proc.of the IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, pp.4375-4380,2018.
[5]T.Kinugasa, K.Hosoda, M.Iribe,F.Asano, Y.Sugimoto,"Special Issue on Dynamically
and Biologically Inspired Legged Locomotion"on Journal of Robotics and
Mechanics,2017.
4.2 《腿型机器人的支撑面评估Support Surface Estimation for Legged Robots》Timon Homberger, Lorenz Wellhausen, Peter Fankhauser, Marco Hutter
了解地形几何是实现腿式机器人安全移动的立足点规划的关键，然而，在可穿透或高度兼容的地形（如草地）上，支撑地面的能见度受到阻碍，即深度传感器无法直接感知。作者通过融合有关脚接触闭合位置的触觉信息和外部感知来估计潜在地形地形。该方法的设计使得它能够提供连续的支撑面估计。即使在相机故障的极端情况下，系统也仅使用本体感觉提供支持面估计。
[1] P. Fankhauser,“Perceptive locomotion for legged robots in rough terrain,” Ph.D. dissertation,
ETH Zurich, 2018.
[2]P. Ross, A. English, D. Ball, B. Upcroft, and P. Corke, “Finding the ground hidden in the
grass: Traversability estimation in vegetation.”
[3]J. Z. Kolter, Y. Kim, and A. Y. Ng, “Stereo vision and terrain modeling for quadruped
robots,” in 2009 IEEE International Conference on Robotics and Automation, May
2009, pp. 1557–1564.
4.3 《力矩可控机器人的ALMA关节运动与操纵ALMA - Articulated Locomotion and Manipulation for a Torque-Controllable Robot》C. Dario Bellicoso, Koen Kramer, Markus St ¨ auble, Dhionis Sako,Fabian Jenelten, Marko Bjelonic, Marco Hutter
机器人在非结构化环境中可能需要与人类协作，本文提出了一种基于六自由度机械臂的扭矩控制四足机器人的运动规划与控制框架ALMA，它能够在执行操纵任务的同时进行动态运动。放松手臂的工作空间。整个系统的扭矩控制实现了柔性行为，允许用户与机器人安全交互。
[1]C. D. Bellicoso, M. Bjelonic, L. Wellhausen, K. Holtmann, F. Gunther, ¨M. Tranzatto,
P. Fankhauser, and M. Hutter, “Advances in real-world applications for legged
robots,” accepted for Journal of Field Robotics,2018.
[2]B. U. Rehman, D. G. Calwell, and C. Semini, “Centaur robots a survey,” in Human-centric
Robotics-Proceedings Of The 20th International Conference Clawar 2017. World
Scientific, 2017, pp.247–258.
4.4《四足机器人多功能动态运动的实时模型预测控制Real-time Model Predictive Control for Versatile Dynamic Motions in Quadrupedal Robots》 Yanran Ding, bhishek Pandala, and Hae-Won Park
作者提出了一种新的模型预测控制（MPC）框架，可以线性化旋转矩阵，而无需借助欧拉角和四元数等参数，从而分别避免了奇点问题和散开现象的问题。
[1]Y Ding, HW Park “Design and Experimental Implementation of a Quasi-Direct-Drive Leg for Optimized Jumping” IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) September 24–28, 2017
[2]Yanran Ding, Chuanzheng Li, Hae-Won Park，“Single Leg
Dynamic Motion Planning with Mixed-Integer Convex Optimization” IEEE/RSJ
International Conference on Intelligent Robots and Systems，2018
[3]Quadrupedal galloping control for a wide range of speed via vertical impulse scaling[J].
Hae-Won Park,Sangbae Kim. Bioinspiration & Biomimetics. 2015
4.5《基于可行脉冲集的步行机器人在线步态转换与干扰恢复Online Gait
Transitions and Disturbance Recovery for Legged Robots via the Feasible Impulse
Set》 Chiheb Boussema, Matthew J. Powell, Gerardo Bledt ,Auke J. Ijspeert, Patrick M. Wensing and Sangbae Kim
四足机器人的步态通常是手动调节的，并且基于时间的。作者提出了一个新概念，可行脉冲集，用于量化四足机器人在线步态出现和适应的腿部能力，无需固定计时或预先确定立足点序列。
[1]Patrick M. Wensing, Luther R. Palmer, David E. Orin "Efficient recursive dynamics
algorithms for operational-space control with application to legged
locomotion"Autonomous Robots, 2015, Vol.38 (4), pp.363-381
4.6《被动行走和跑步：ATRIAS实况演示的经验教训Walking and Running with Passive Compliance: Lessons from Engineering a Live Demonstration of the ATRIAS Biped》 Christian Hubicki, Andy Abate, Patrick Clary, Siavash Rezazadeh, Mikhail Jones, Andrew Peekema ,Johnathan Van Why,
Ryan Domres, Albert Wu, William Martin ,Hartmut Geyer , and Jonathan Hurst
长期以来，生物两足动物一直被认为利用柔顺性和被动动力学来行走和奔跑，但是在机器人运动实践中很难实现这种方式。Atrias是一种双足机器人，设计目的是利用由调谐的机械顺应性而产生的固有的稳定效应。它结合了工程被动动力学与互补控制算法，能够在没有外部电源的情况下通过被动动力行走和奔跑或者支持。
[1]Christian Hubicki, Jesse Grimes, Mikhail Jones, Daniel Renjewski,Alexander Sprowitz;,Andy
Abate，Jonathan Hurst,“ATRIAS: Design and
validation of a tether-free 3D-capable spring-mass bipedal robot”,The
International Journal of Robotics Research,SAGE Publications,2016,pp.1497-1521
Legged Robots V
5.1 Self-Modifying
Morphology Experiments with DyRET: Dynamic Robot for Embodied Testing*
Tønnes F. Nygaard , Charles P. Martin ,
Jim Torresen and Kyrre Glette
本文提出了一种自我重构形态的方式允许机器人调整腿的长度以适应不同的环境。共四条腿，每条腿有五个自由度，股骨和胫骨的两个下连杆都可以被重新配置为不同的长度。基于ROS系统。实验室结果：高电压下，步态延长，长腿的机器人效果更好；低电压下，短腿的机器人效果更好。车库里，基础步态0.9m/min；长腿延长步态达到1.2m/min。
[1] T. F. Nygaard, E. Samuelsen, and K. Glette, “Overcoming initial convergence in multi-objective evolution of robot control and morphology using a two-phase approach,” in 2017 Applications of Evolutionary Computation (EvoStar17), G. Squillero and K. Sim, Eds. Springer
International Publishing, 2017, pp. 825–836
[2] T. F. Nygaard, C. P. Martin, E. Samuelsen, J. Torresen, and K. Glette, “Real-world evolution adapts robot morphology and control to hardware limitations,” in 2018 Genetic and Evolutionary
Computation Conference (GECCO18). ACM, 2018, pp. 125–132.
[3] T. F. Nygaard, “DyRET GitHub
repository.” [Online]. Available: https://github.com/dyret-robot/dyret documentation
[4] J. Nordmoen, K. O. Ellefsen, and K. Glette, “Combining mapelites and incremental evolution to generate gaits for a mammalian quadruped robot,” in Applications of Evolutionary Computation, K. Sim and P. Kaufmann, Eds. Cham: Springer International Publishing, 2018, pp. 719–733.
[5] T. F. Nygaard, J. Torresen, and K. Glette, “Multi-objective evolution of fast and stable gaits on a physical quadruped robotic platform,” in 2016 IEEE Symposium Series on Computational
Intelligence (SSCI), 2016, pp. 1–8.
5.2 Experimental Validation
of High-Efficiency Hydraulic Direct-Drive System for a Biped Humanoid Robot —Comparison with Valve-Based Control System
J. Shimizu, T. Otani, H. Mizukami, K. Hashimoto, and A. Takanishi, Member, IEEE
本文提出了一个名叫WABIAN-2R的配有盆骨模型并且可以执行伸展的膝盖步态的机器人。利用液压直驱系统两条腿共享电机输出来减小电机的尺寸，实现尽可能多的功能。具有优秀的节能性、速度跟随以及近乎完美的位置追踪。
[1] Y. Ogura, K. Shimomura, H. Kondo, A. Morishima, T. Okubo, S. Momoki, H. O. Lim and A. Takanishi, “Human-like Walking with Knee Stretched, Heel-contact and Toe-off Motion by a Humanoid robot,” Proc. 2006 IEEE/RSJ Int. Conf. Intelligent Robots and Systems, pp. 3976-3981, 2006.
[2] T. Otani, K. Hashimoto, S. Hamamoto, S. Miyamae, M. Sakaguchi, Y. Kawakami, H.O. Lim and A. Takanishi, “Knee Joint Mechanism That Mimics Elastic Characteristics and Bending in Human Running,” Proc. 2015 IEEE/RSJ Int. Conf. Intelligent Robots and Systems, pp. 5156-5161, 2015.
[3] J. Shimizu, T. Otani, K. Hashimoto, A. Takanishi, “Downsizing the Motors of a Biped Robot Using a Hydraulic Direct Drive System,” Proc. 2018 IEEE-RAS Int. Conf. on Humanoid Robots, pp. 580-586, 2018.
5.3 Experimental Demonstration of High-Performance Robotic Balancing
Josephus J. M. Driessen, Antonios E. Gkikakis , Roy Featherstone and B. Roodra P. Singh
本文对近来的一种平衡控制理论进行了首次验证，该理论提出允许进行机器人进行大而快的运动时保持高性能的平衡。但是仅仅验证了2D层面，3D层面还没有被解决。
[1] M. Azad and R. Featherstone, “Angular momentum based balance controller for an under-actuated planar robot,”
Autonomous Robots, vol. 40, no. 1, pp. 93–107, 2016.
[2] J. J. M. Driessen, R. Featherstone, and A. E. Gkikakis, “An actuator design criterion to maximize physical balance recovery,” in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Madrid, Spain, October 2018.
[3] R. Featherstone, “Quantitative Measures of a Robot’s Ability to Balance,” in Proceedings of Robotics: Science and Systems, Rome, Italy, July 13–17, 2015. [Online]. Available: http://www.
roboticsproceedings.org/rss11/p26.html
5.4 Design of Anti-skid Foot with Passive Slip Detection Mechanism forConditional Utilization
of Heterogeneous Foot Pads
Jaejun Park, Do Hun Kong and Hae-Won Park
本文介绍了一种新颖的防滑足部，足部由橡胶制成的主足和由线性约束脊柱机制和锚固脊柱形成的副足组成。其所使用的被动滑动检测机制和机构特殊的锁释机制能够保证在主足打滑时使用副足。此种防滑足部在混凝土、木制和沙土地面上能承受的最大切向载荷分别是普通橡胶足部的190%，165%，275%。采用此种足部的两组平面机器人能够跃上50°的斜坡，并且能够在最大倾角为61.5°的混凝土斜面上保持站立。
[1] Y. Ding and H.-W. Park, “Design and experimental implementation of a quasi-direct-drive leg for optimized jumping,”
in IEEE/RSJ International Conference on Intelligent Robots and Systems, Sept
2017, pp. 300–305.
[2] H.-W. Park, S. Park, and S. Kim,
“Variable-speed quadrupedal bounding using impulse planning: Untethered
high-speed 3d running of mit cheetah 2,” in Robotics and automation (ICRA),
2015 IEEE international conference on. IEEE, 2015, pp. 5163–5170.
[3] H.-W. Park, P. M. Wensing, and S. Kim,
“High-speed bounding with the mit cheetah 2: Control design and experiments,”
The International Journal of Robotics Research, vol. 36, no. 2, pp. 167–192,
2017
[4] H.-W. Park, P. Wensing, and S. Kim,
“Online planning for autonomous running jumps over obstacles in high-speedquadrupeds,” in Proceedings of Robotics: Science and Systems, Rome, Italy, July 2015
5.5 Trajectory
Optimization for Wheeled-Legged Quadrupedal Robots using Linearized ZMP Constraints. Yvain de Viragh,Marko Bjelonic, C. Dario Bellicoso, Fabian Jenelten, and Marco Hutter
本文展示了为带驱动轮的四足机器人所设计的一种轨迹优化器。通过级联的形式求解基体轨迹和足轨迹的角、垂直和竖直分量，并通过引入零点平衡标准的新线性公式，该方法仅依赖于二次编程，从而消除了非线性优化例程的需要。这是第一次使用在线运动规划器控制轮足式机器人的形态运动。
[1] M. Hutter et al., “ANYmal – toward legged robots for harsh environments,” Adv. Robot., vol. 31, pp. 918–931, 2017.
[2] A. W. Winkler, C. D. Bellicoso, M. Hutter, and J. Buchli, “Gait and trajectory optimization for legged systems through phase-based endeffector parameterization,” IEEE Robot. Autom. Lett.,
vol. 3, no. 3, pp. 1560–1567, Jul. 2018.
[3] C. D. Bellicoso, F. Jenelten, P. Fankhauser, C. Gehring, J. Hwangbo, and M. Hutter, “Dynamic locomotion and whole-body control for quadrupedal robots,” in Proc. IEEE/RSJ Int. Conf. Intell. Robots Syst., Sep. 2017, pp. 3359–3365.
[4] C. D. Bellicoso, F. Jenelten, C. Gehring, and M. Hutter, “Dynamic locomotion through online nonlinear motion optimization for quadrupedal robots,” IEEE Robot. Autom. Lett., vol. 3, no.
3,pp. 2261–2268, Jul. 2018.
