Non-reciprocating legged gait for robot with epicyclic-gear-based eccentric paddle mechanism

https://doi.org/10.1016/j.robot.2015.02.004Get rights and content

Highlights

  • A non-reciprocating legged gait is proposed for an ePaddle-EGM-based robot.

  • Gait sequence and planning method for this gait are discussed and verified.

  • Locomotion accuracy is secured with the gait in the presence of gear backlash.

  • Robot walks more efficiently in this gait than in legged crawling gait.

Abstract

A novel eccentric paddle mechanism based on epicyclic gear mechanism (ePaddle-EGM) has been proposed to enhance the mobility of amphibious robots for multi-terrain tasks with diverse locomotion gaits. This paper proposes a unique non-reciprocating legged walking gait for a quadruped or hexapod robot based on ePaddle-EGMs after a brief kinematic analysis of the ePaddle-EGM. During the gait, the robot can generate desired periodic motion of the legs in the swinging and supporting phases while maintaining all the actuators rotating unidirectionally. Advantages of the proposed non-reciprocating legged gaits include that influence of backlash in the epicyclic gear mechanism could be eliminated and accuracy of locomotion can be guaranteed. Experiments on a single ePaddle-EGM prototype module were conducted to validate the proposed mechanism design and the idea of deploying the non-reciprocating legged walking gait to improve locomotion accuracy and energetic efficiency of the robot.

Introduction

In recent years, a lot of effort has been put into developing autonomous amphibious robots for performing high-risk tasks in harsh amphibious environments, e.g. mine clearing, rescue and search after disaster, terrain mapping, and scouting potential approach lanes. Typical amphibious environments are usually covered by terrestrial, aquatic, or a combination of both. Amphibious robots deployed in such environments should have the capability to operate both on water and land with multi-terrain mobility.

There exist several amphibious robots, which fall into two categories: biomimetic amphibious robots and hybrid-mechanism-based amphibious robots. The former is inspired by the morphological features of natural amphibious creatures. Robots that belong to this category are snake-like robot ACM-R5 by Hirose  [1], salamander robot AmphiBot and Salamandra Robotica II by Crespi  [2], [3], fish-like amphibious robot AmphiRobot-II by Yu  [4], lobster robot by Ayers  [5], and, turtle mimicking robot MiniTurtle-I by Chen  [6]. These amphibious robots try to mimic the biologic characteristics and functions of their biological counterparts  [7]. However, as their biological counterparts, biomimetic amphibious robots also have limitations on their working environments. For example, slender snake robots, fish-like robots, salamander robots, and robotic turtles are capable of moving both in water and on land, but they might get stuck in wetland or on rough terrains. The lobster robot can walk on the land or the sea bottom, but it cannot swim in the water.

In contrast with biomimetic amphibious robots, amphibious robots based on hybrid mechanism achieve the amphibious mobility by integrating several basic motion units into one locomotion mechanism, such as the wheel-leg-tail-integrated robot DAGSI Whegs by Boxerbaum  [8], paddle-wheel integrated autonomous amphibious vehicle developed by Frejek  [9], leg-paddle-integrated robot AQUA  [10], and propeller-leg-integrated spherical amphibious robot by Shi  [11]. Most hybrid mechanism robots employ two or more sets of propulsion mechanisms to achieve terrestrial and aquatic locomotion. This set-up results in a bulky body and redundant mechanisms which compromise energetic efficiency of the locomotion. Motion planning and controlling of these robot are complicated as well.

Although numerous achievements have been made in amphibious robot design, amphibious robots remain inadequate for practical tasks, for instance, search and rescue tasks after the occurrence of a tsunami or flood. In this situation, it is desired that amphibious robots can travel on uneven ground, swim in water, pass through transitional zones such as sandy beach and muddy pool, and with payloads. Most existing amphibious robots are difficult to apply in these practical high-risk tasks, due to the limitation of mobility, motion efficiency and load capability.

Based on these considerations, we start our effort to develop an amphibious robot that possesses high mobility and high energetic efficiency in complex amphibious environment by proposing a novel locomotion mechanism: eccentric paddle mechanism (ePaddle)  [12]. The concept of the ePaddle is shown in Fig. 1(a). By actively locating the paddle shaft inside the wheeled shell via independent actuators, motion patterns of the ePaddle can be alternated. Five locomotion gaits (as shown in Fig. 1(c)) have been experimentally validated, including the wheeled rolling gait  [13], the legged-walking gait  [14], the legged-wheeled-hybrid gait  [15], the rotational paddling gait  [16] and the oscillating paddling gait  [17], [18]. The wheeled gait has the best efficiency and suitable for fast traveling on even ground. The legged-walking gait has lower energetic efficiency but allows the robot to walk on rough terrain. On soft terrain where the robot slips or sinks, the legged-wheeled-hybrid gait can be used. Though the rotational paddling and the oscillating paddling gaits have worst energetic efficiency among all the possible gaits, they enable an ePaddle-based robot to swim underwater. The ability to generate thrusts for underwater propulsion have been experimentally proven  [17], [18].

Performance comparison between the ePaddle-based robot and other amphibious robots is still ongoing. However, high mobility of an ePaddle-based robot in complex amphibious environment is expected, because an ePaddle-based robot has a better chance to cross challenging terrains by switching among its versatile gaits, as opposed to other robots with a limited number of gaits. The robot can also exhibit high energy efficiency by choosing the relatively high efficient gait according to the terrain rather than trying to adopt a universal gait on all the terrains.

Two prototypes of the ePaddle mechanism have been designed, in which a slide screw mechanism (as shown in Fig. 1(b)) and an epicyclic gear mechanism (as shown in Fig. 2(b)) have been respectively used for positioning the paddle shaft  [15], [19]. The prototype based on epicyclic gear mechanism (ePaddle-EGM) has advantages in reliability and energy efficiency, but the accuracy of the paddles motion is easily jeopardized by backlash in the gear train. The legged gait is an example gait that significantly relies on motion accuracy of paddles tip; the larger the backlash the greater the chance that the robot loses its walking stability.

Traditional anti-backlash methods, such as equipping the gear train with specially designed mechanical backlash-elimination devices and programmatic compensating backlash in motion planning algorithms, complicate the design, control, and maintenance of the ePaddle mechanism. In this paper, we present a simple kinematics-based motion planning strategy for eliminating the impact of the backlash of the ePaddle-EGM in the legged walking gait. This strategy is inspired by a unique dynamic coupling property of the ePaddle-EGM: the planar motion of the paddle shaft is result of two coupled rotational motions of the planetary gear and the carriage. Following this strategy, the periodically reciprocated motion of the paddles tip required in legged walking gait can be generated by non-reciprocating motion of the motors. Since all the motors and gears are rotating in one direction in the walking cycle, the impact of backlash vanishes and motion accuracy is guaranteed even backlash in the gear train increases as the gear wears. Besides, non-reciprocating motion of the motors further improves the energetic efficiency of the ePaddle-EGM, since motors are far more efficient running in a single direction with velocity bounds.

The rest of the paper is organized as follows. Section  2 describes the design of the ePaddle prototype based on epicyclic gear mechanism, including kinematic analysis and mechanism parameter selection. Section  3 presents the motion planning method for achieving efficient and precise non-reciprocating legged-walking gait. Section  4 discusses the experimental results and finally conclusions are drawn in Section  5.

Section snippets

Conceptual design of the ePaddle-EGM

The proposed new version of ePaddle is named ePaddle-EGM, which is the abbreviation for eccentric paddle mechanism based on epicyclic gear mechanism. Four or six ePaddle modules can compose a quadruped or hexapod ePaddle-based robot. As shown in Fig. 2, ePaddle-EGM is composed of five main components: (1) a paddle shaft that can be actively positioned via an epicyclic gear mechanism, (2) a set of four paddles that can passively rotate around the paddle shaft, (3) a wheel-like shell that can be

Planning of non-reciprocating legged gait

The epicyclic gear mechanism in ePaddle-EGM brings several advantages, such as higher reliability, higher efficiency, and dynamic coupling. However, a disadvantage is brought by this mechanism, which is the inaccuracy of the motion due to gear backlash. The backlash has a significant impact on the precision of paddle shafts trajectory. Failure to take accurate action could substantially degrade the performance of locomotion gait, especially the legged walking gait.

The legged walking gait is the

Experiment setups

Since the prototype robot with multiple ePaddle-EGM modules has not been built yet, we conducted the experiments on a single ePaddle-EGM module to verify the idea of non-reciprocating legged gait. The accuracy and energetic efficiency of the non-reciprocating legged gaits are evaluated with experiment setups as shown in Figs. 9(a) and (b), respectively.

The setup shown in Fig. 9(a) includes the ePaddle-EGM prototype, a realtime control system, and a motion capture system. The control system

Conclusion

In this paper, an eccentric paddle mechanism based on epicyclic gear mechanism has been presented for amphibious robots. Compared with the previous slide-screw-based ePaddle mechanism, this ePaddle-EGM module features higher reliability, better efficiency, and dynamic coupling. A novel non-reciprocating legged walking gait for a quadruped or hexapod robot based on ePaddle-EGM modules has been proposed. The distinguishing feature of this novel legged gait is that the periodic reciprocating

Acknowledgments

This work was supported by the National Natural Science Foundation of China  [grant numbers 61305127, 61233010, and 61203348]; the Creative Research Fund of Shanghai University[grant number sdcx2012019]; and the University Young Teachers Training Scheme of Shanghai.

Huayan Pu received her M.Sc. and Ph.D. degrees in Mechatronics Engineering from Huazhong University of Science and Technology (Wuhan, China), in 2007 and 2011, respectively. Since 2011, she has been a Lecturer at Shanghai University (Shanghai, China). She was awarded the best paper in Biomimetics at 2013 IEEE International Conference on Robotics and Biomimetics. She was also nominated the best conference paper finalist at 2012 IEEE International Conference on Robotics and Biomimetics. Her

References (20)

There are more references available in the full text version of this article.

Cited by (9)

  • Revelation of metamorphic phenomenon through the equivalent mechanisms and development of the novel metamorphic epicyclic gear trains

    2021, Mechanism and Machine Theory
    Citation Excerpt :

    Several graph-theoretic methods [31-34] have been developed to simplify the kinematic analysis of gear trains. In reference [35,36], an ePaddle-EGM robot was designed and optimized based on the relative position of the sun gear and planetary gear in an epicyclic gear train. Laus et al. [37] analysed the efficiency of epicyclic gear trains using graph and screw theories.

  • Optimized non-reciprocating legged gait for an eccentric paddle mechanism

    2018, Robotics and Autonomous Systems
    Citation Excerpt :

    Traditional anti-backlash methods, such as equipping the gear train with specially designed mechanical backlash-elimination devices and programmatic compensating backlash in motion planning algorithms, complicate the design, control and maintenance of the mechanism. In the previous study, a novel gait named non-reciprocating legged gait has been proposed [16]. In which, all the actuators rotated in one direction without reciprocating during the walking, and the performance improvements in locomotion accuracy and energetic efficiency have been verified by the experiments.

  • Evaluation of Energy Efficiency of an Eccentric Paddle Mechanism on Sandy Terrain

    2020, 2020 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2020
View all citing articles on Scopus

Huayan Pu received her M.Sc. and Ph.D. degrees in Mechatronics Engineering from Huazhong University of Science and Technology (Wuhan, China), in 2007 and 2011, respectively. Since 2011, she has been a Lecturer at Shanghai University (Shanghai, China). She was awarded the best paper in Biomimetics at 2013 IEEE International Conference on Robotics and Biomimetics. She was also nominated the best conference paper finalist at 2012 IEEE International Conference on Robotics and Biomimetics. Her current research interests include modeling, control, and simulation of field robotics and locomotion system.

Jinglei Zhao received his B.Eng. degree in Mechanical Engineering from Shanghai University (Shanghai, China) in 2012. Since 2012, he has been a graduate student at Shanghai University (Shanghai, China). His research interests include modeling, control, and simulation of field robotics and locomotion system.

Yi Sun received his B.Eng. degrees in both Mechanical Engineering and in Computer Science and Technology from Huazhong University of Science and Technology (Wuhan, China). He then went on to receive his M.Eng. degree in Mechatronic Engineering from Huazhong University of Science and Technology in 2006. He received his Dr. Eng. degree in Robotics from Ritsumeikan University (Kyoto, Japan) in 2013. He is currently a Senior Researcher at Ritsumeikan University. He was awarded the best paper in Biomimetics at 2013 IEEE International Conference on Robotics and Biomimetics. He was also nominated the best conference paper finalist at 2012 IEEE International Conference on Robotics and Biomimetics. His research interests include artificial locomotion systems and amphibious robots.

Shugen Ma received his Dr. Eng. degree in Mechanical Engineering Science from Tokyo Institute of Technology (Tokyo, Japan) in 1991. From 1991 to 1992 he worked for Komatsu Ltd as a Research Engineer, and from 1992 to 1993 he was at the University of California (Riverside, USA) as a visiting scholar. Since July 1993 he has been with Ibaraki University (Japan) as an Associate Professor of the Department of Systems Engineering. In October 2005, he joined Ritsumeikan University as a professor in the Organization for Promotion of the COE program and currently is a Professor in the Department of Robotics. He is also holding Professor position at Tianjin University and a “Ziqiang” Professor of Shanghai University. His current research interests include the design and control theory of novel robots, rescue robotics, and Bio-robotics.

Jun Luo is a Professor in the School of Mechatronics Engineering and Automation of Shanghai University, China. He is the head of Precision Mechanical Engineering Department at the Shanghai University and the Vice Director of Shanghai Municipal Key Laboratory of Robotics. He received his B.S. degree in Mechanical Engineering from Henan Polytechnic University, Jiaozuo, China, in 1994, M.S. degree in Mechanical Engineering from the Henan Polytechnic University, Jiaozuo, China, in 1997, and Ph.D. degree from the Research Institute of Robotics in Shanghai Jiao Tong University, Shanghai, China, in 2000. His research areas include robot sensing, sensory feedback, mechatronics, man–machine interfaces, and special robotics.

Zhenbang Gong is a Professor in the School of Mechatronics Engineering and Automation of Shanghai University, China. He received his B.S. degree in Mechanical Engineering from Shanghai Science and Technology University (Shanghai, China) in 1964. He is the Director of the Precision Machinery Institute, Shanghai University, China, the Chairman of Expert Committee of Robotics Union, Shanghai of China, and also was the Member (Contact Person of China) of Joint Coordinating Forum of the International Advanced Robotics Program. His research areas include robot sensing, sensory feedback, mechatronics, man–machine interfaces, and special robotics.

View full text