Jumping efficiency of small creatures and its applicability in robotics

Available online at www.sciencedirect.com Available online at www.sciencedirect.com

ScienceDirect ScienceDirect

Procedia Manufacturing 00 (2017) 000–000 Available Availableonline onlineatatwww.sciencedirect.com www.sciencedirect.com Procedia Manufacturing 00 (2017) 000–000

ScienceDirect ScienceDirect 

www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia

Procedia Manufacturing 21 (2018) 243–250 Procedia Manufacturing 00 (2017) 000–000 www.elsevier.com/locate/procedia

15th Global Conference on Sustainable Manufacturing 15th Global Conference on Sustainable Manufacturing

Jumping efficiency of small creatures and its applicability in Jumping efficiency of small creatures and its applicability in robotics Manufacturing Engineering Society International Conference 2017, MESIC 2017, 28-30 June robotics 2017, Vigo (Pontevedra), Spain Uri Ben Hanan**, Avi Weiss, Valentin Zaitsev Uri Ben Hanan , Avi Weiss, Valentin Zaitsev

ORT Braude Mechanicaloptimization Engineering, Snunit Str. 51, 2161002, Israel Costing models for College, capacity inKarmiel Industry 4.0: Trade-off ORT Braude College, Mechanical Engineering, Snunit Str. 51, Karmiel 2161002, Israel between used capacity and operational efficiency

Abstract Abstract A. Santanaa, P. Afonsoa,*, A. Zaninb, R. Wernkeb Among different motion gaits, jumping is a quick and efficient locomotion for advancing in rough, uneven or unpredicted Among jumpingAs a quick and efficient locomotion for advancing rough, unevenarea, or unpredicted a is the terrains, different used by amotion varietygaits, of animals. muscle power generation is proportional to itsincross-sectional small-scale University of Minho, 4800-058 Guimarães, Portugal bthetiny terrains, used by a variety animals.inAs muscle power proportional to itscreated cross-sectional area, small-scale jumpers suffer from power of limitation their muscles. Togeneration overcome this barrier, special mechanisms inside Unochapecó, 89809-000 Chapecó,is SC, Brazil nature jumpers suffer that fromare power in their tiny muscles. Toranges overcome barrier, nature created special mechanisms inside these jumpers able limitation to propel them to very impressive withthis respect to their own size. The governing principle is these jumpers able elements, to propel them very impressive ranges itwith respect as to their own Researchers size. The governing is loading energythat intoare elastic and attothe right time releasing as quickly possible. designingprinciple miniature loading energy into elastic elements, and at the right time releasing it as face quickly as possible. Researchers designing artificial jumping mechanisms for improving small-scale robots mobility, similar power limitations of motors and miniature batteries. artificial mechanisms for small-scale improving small-scale mobility, similar power limitations of motors and Abstract The very jumping same principles used by jumpers arerobots implemented in face the bio-inspired artificial mechanisms, and arebatteries. utilized The very same principles used by small-scale jumpers are implemented in the bio-inspired artificial mechanisms, and are utilized in jumping robots. In this work, a jumping robot implementing bio-inspired principles, is presented. in jumping In this a jumping implementing bio-inspired Under the robots. concept of work, "Industry 4.0",robot production processes will principles, be pushedis presented. to be increasingly interconnected, © 2017 The Authors. Published by Elsevier B.V. necessarily, much more efficient. In this context, capacity optimization information based on a real time basis and, © 2017 2018 The Published Elsevier B.V. © The Authors. Authors. Published by by Elsevier B.V.maximization, Peer-review under responsibility ofof thecapacity scientific committee of the 15th Globalalso Conference on Sustainable Manufacturing. goes beyond the traditional contributing for organization’s profitability and value. Peer-review under responsibilityaim of the scientific committee of the 15th Global Conference on Sustainable Manufacturing (GCSM). Peer-review under responsibility of the scientific of the 15th Global Conference on Sustainable Manufacturing. Indeed, lean management and continuous committee improvement approaches suggest capacity optimization instead of

Keywords: Jumping robots; Jumping animals; bio-inspiered maximization. The study of capacity optimization and costing models is an important research topic that deserves Keywords: Jumping robots; Jumping animals; bio-inspiered contributions from both the practical and theoretical perspectives. This paper presents and discusses a mathematical model for capacity management based on different costing models (ABC and TDABC). A generic model has been 1. Introduction developed and it was used to analyze idle capacity and to design strategies towards the maximization of organization’s 1. Introduction value. The trade-off capacity maximization vs operational efficiency is highlighted and it is shown that capacity The need to improve robots mobility is evident in recent years. Robots are required to reach difficult areas, while optimization might hide operational inefficiency. The need to improve robots mobility is grounds. evident inUneven recent years. Robots are required to reach difficult areas, while travelling in rough terrains, such as rocky surfaces, obstacles that might appear in the robot's path, © 2017 The Authors. Published by Elsevier B.V. travelling in rough terrains, such as rocky grounds. Uneven surfaces, obstacles that might appear in the robot's path, the need to reach high places and many additional uncertainties, are expected to challenge the robot's mobility [1]. Peer-review under responsibility of the scientific committee of the Manufacturing Engineering Society International Conference the need to reachofhigh places and many are expected challenge robot's mobilitywith [1]. Jumping is one the efficient ways for additional locomotionuncertainties, of mobile robots [2], as ittoreduces the the overall interaction 2017. Jumping is while one ofmoving, the efficient ways for locomotion of mobile robots as itreach reduces thepositions. overall interaction with the ground enables to overcome obstacles, and helps the [2], robots higher It is especially the ground while moving, enables to overcome obstacles, and helps the robots reach higher positions. It is especially importantCost to maneuver when the Capacity obstacles are large Idle withCapacity; respectOperational to the robot's size [3]. Keywords: Models; ABC; TDABC; Management; Efficiency important to maneuver when the obstacles are large with respect to the robot's size [3]. * Corresponding author. Tel.: 972-4-9901830; fax: 972-4-9901886.

1. Introduction * E-mail Corresponding Tel.: 972-4-9901830; fax: 972-4-9901886. address:author. [email protected] E-mail address: [email protected]

The cost of idle capacity is a fundamental information for companies and their management of extreme importance in modern systems. In general, it isB.V. defined as unused capacity or production potential and can be measured 2351-9789 ©production 2017 The Authors. Published by Elsevier 2351-9789 2017responsibility The Authors. Published by Elsevier B.V.hours Peer-review of the scientific committee of the 15th Global Conference on Sustainable Manufacturing. in several©under ways: tons of production, available of manufacturing, etc. The management of the idle capacity Peer-review underTel.: responsibility the761; scientific committee the 15th Global Conference on Sustainable Manufacturing. * Paulo Afonso. +351 253 of 510 fax: +351 253 604of741 E-mail address: [email protected]

2351-9789 © 2017 The Authors. Published by Elsevier B.V. Peer-review under responsibility of the scientific committee of the Manufacturing Engineering Society International Conference 2017. 2351-9789 © 2018 The Authors. Published by Elsevier B.V. Peer-review under responsibility of the scientific committee of the 15th Global Conference on Sustainable Manufacturing (GCSM). 10.1016/j.promfg.2018.02.117

Uri Ben Hanan et al. / Procedia Manufacturing 21 (2018) 243–250 Uri Ben Hanan / Procedia Manufacturing 00 (2017) 000–000

244

1

Previous studies have thus found jumping to be the most suitable form of locomotion for mobile robots traversing problematic terrain [4, 5 and 6]. The animal kingdom has many jumping animals, in particular insects, which demonstrate impressive jumps with respect to their small size. Therefore, Nature serves as a source for inspiration to develop artificial jumping mechanisms [4]. Indeed, many nature-inspired jumping robots have been developed thus far (see section 1.2). Nevertheless, with the current technology and mechatronics, researchers are still far from the point that an artificial mechanism could be compared with animals' jumping mechanisms. The main limitation is the power limit of available motors and batteries. While the most promising designs rely on electric motors, gears, and cam mechanisms, their power enhancement usually comes with the expense of extra weight and volume. Therefore, such cam-based mechanisms are limited by the energy they can capacitate with respect to their size and weight. This paper offers a solution to this problem, by using available mechatronics combined with insect-inspired legged jumping structure. 1.1 Jumping in Nature In nature, there are many animals that use jumping in order to improve their mobility and maneuverability, usually as a secondary gait after walking and running. Animals often use the jump for quick escape from predators, or as a means for pre-flight locomotion gaits such as gliding or flying. In small-scale jumpers, the jump is a critical maneuver, since all surrounding objects and obstacles are large with respect to the animal size, limiting its other possible gaits. In addition, the jump enables a quick change in position using a relatively small amount of energy. There is however a main difference between large-scale jumpers (e.g., kangaroos, klipspringers, hares, humans etc.) and small-scale jumpers such as jumping insects (e.g., fleas, grasshoppers, froghoppers, locusts etc.). Whereas largescale jumpers have large muscles with the ability to generate high power, small-scale jumpers have small muscles with limited power, since the power generation ability is proportional to the muscle's cross-sectional area. As the scale of the animal decreases, its cross-sectional area of muscles decreases as well, limiting its power generation abilities. To overcome these limitations, small-scale jumpers incorporate mechanisms that are able to propel these creatures to very impressive ranges with respect to their own size. The principle that lies in the heart of all these mechanisms lies at the initial preparation stage before the leap itself. Small animals load energy into elastic elements located in their jumping mechanisms, and at the right time, they convert this elastic energy very fast into kinetic energy, and propel themselves into the air. Slowly accumulating and storing energy circumvents the muscles power limitation in small-scale jumpers, accumulating large amounts of energy with respect to their own weight. As a result, they can catapult themselves to large distances with respect to their own size. Table 1 presents typical smallscale jumpers' performance, including their typical mass and size, kinematics, and actuation method, as available in the literature. The jumping range to mass ratio was not calculated; as such comparison is inaccurate, since the creatures have dozens of species that vary in mass and size. Table 1. Jumping performance of common jumping insects. Insect

Mass (mg)

Size (mm)

Jumping height (cm)

Jumping distance (cm)

Jumping energy (mJ)

Take-off speed (m/s)

Acceleration (m/s2)

Take-off period (ms)

Actuation

Flea [7]

0.45

1.5

4.9

-

0.225

1

1330

0.75

Elastic+legs

Spider [8]

10

6

-

-

-

0.67

0.0513

-

Hydraulic+legs

Froghopper [9] Grasshopper [10,11] Click beetle [12] Locust [13]

12.3

6.1

43

-

0.136

4.7

5400

0.875

Elastic+legs

1400

-

25

35

2.5

3

180

-

Elastic+legs

29

11

25

-

0.115

2.2

-

0.6

Elastic+pivot

2500

55

-

100

-

3.2

180

25

Elastic+legs

2

Uri Ben Hanan et al. / Procedia Manufacturing 21 (2018) 243–250 Uri Ben Hanan / Procedia Manufacturing 00 (2017) 000–000

245

Although the spider's jumping mechanism relies on hydraulics, and the click beetle's jumping mechanism relies on rotational pivot axle, the more common jumping mechanisms in insects rely on large powerful jumping legs. The legs are initially folded in a closed position. After the energy is stored in special elastic cuticle located in the legs' joints, by means of powerful muscle contraction, this energy is quickly released in the result of sudden muscle relaxation. This causes the legs to quickly extend, providing the take-off speed for the animal's body. From Table 1, it is clear that the lighter the animal, the further it can jump with respect to its own size. With available mechatronic components, it is a challenging task to recreate a jumping mechanism the scale of insects. However, although current mechatronic devices are limited in mass, the benefits of possible mass and size reduction are clear. The desert locust is the largest and heaviest animal in Table 1. Nevertheless, it has an impressive jumping distance that may be compared with a small-scale robot. Since the locust's dimensions are in the scale of miniature robots, the locust is chosen as the source for inspiration to design a jumping robot. A robot of the same size-scale as the locust would be small enough to perform different tasks while being large enough to enable its manufacture using inexpensive and widely available technologies, such as commercial micro-controllers and miniature electrical motors. In addition, the biological and mechanical aspects of the locust jump have been studied extensively, e.g., [13]. Finally, in addition to jumping, the locust has the ability to perform many other manoeuvres such as righting, walking, gliding and, of course, flying. Consequently, it is an inspiring model for future development of miniature robots whose mobility will be enhanced by integrating additional abilities. Although the locust is the model for designing an artificial jumping mechanism, the aim is creating a robot with a much greater jumping ability than the locust, enabling it to surpass the performance of other jumping mechanisms (see section 1.2). 1.2 Robotic Jumping Mechanisms Locomotion on rough, uneven or unpredicted terrain requires more than just jumping. In addition to the jumping height, other requirements may be self-righting, steering, leap angle control and jumping energy control. Although these features are important, the scope of this paper is extending the jumping height. Therefore, the focus is on the jumping mechanism and its efficiency. The following aims at surveying various existing jumping mechanisms, indicating their jumping heights. Some of the mechanisms developed to date (see table 3 in section 3) are biologically inspired by all kinds of nature’s jumpers, such as fleas and locusts, while others are not necessarily focused on nature as a prime source of locust-inspired jumping robot that reaches a height of about inspiration. Researchers in [14] developed a , and relies on linear springs. It uses a gear system for energy storage, a cam mechanism for energy release, and two simple parallel single links with a flat base to propel the robot from the ground. Another locust-inspired jumping mechanism was presented in [15]. This mechanism mimics the locust jumping leg morphology, using a basic crank-slider mechanism with an elastic element for storing the energy (no prototype was reported). The . It utilizes a torsion spring loaded through a gear version of the EPFL jumper [16], reaches a height of about system for energy storage, a cam mechanism for energy release, and a four-bar mechanism for translating the released energy into a jump. However, in order for this robot to jump, it must be held manually until its springs are charged up to a certain level, then it can be placed back on the ground and finish the jump. The more recent version of this robot solves this drawback. It incorporates self-recovery capability and jumps to a height of around MSU jumper [18] [17], less than the original version due to the extra weight of the righting mechanism. The also utilizes a combination of torsion springs loaded through a gear system for energy storage. However, instead of using a cam mechanism, it uses a cable, pulley, and a rotation link as a release mechanism. Its energy release system , and has comprises a six-bar mechanism imitating two jointed legs. This robot jumps to a height of about additional capabilities, such as steering and self-righting before the jump. A flea-inspired catapult mechanism was demonstrated in [19]. This miniature jumping robot uses a shape memory and jumps as high as . Another interesting approach alloy (SMA) spring as an actuator, weighs only jump. using a micro jumping robot propelled by chemical energy was demonstrated by [20], and achieved An interesting example of a flea-inspired jumper is the sand-flea robot by Boston Dynamics [21], which jumps with an ability to alter the leap angle. This robot, however, is much larger than demonstrates impressive the class of robots mentioned earlier, and its activation is pneumatic.

Uri Ben Hanan et al. / Procedia Manufacturing 21 (2018) 243–250 Uri Ben Hanan / Procedia Manufacturing 00 (2017) 000–000

246

3

Jumping mechanisms designs that are more sophisticated from a mechanical viewpoint rely on electromechanical clutch mechanisms. When the clutch engages, it enables to deliver power from the motor to the springs loading the required jumping energy. When the clutch disengages, it separates the motor's axis from the jumping mechanism, enabling the springs to quickly release the stored energy, and to catapult the robot into the air. The robot and jumps to a height of . Another example is a jumping and presented in [4] uses a clutch, weighs , and it jumps to a height of only , but nevertheless gliding robot [22], which also uses a clutch. It weighs demonstrates a significantly larger range, and lower cost-of-transport than a comparable ballistic jumper. The lightest jumping mechanisms use chemical energy or SMAs for executing the jump. However, these solutions are usually use electrical motors, gears and not capable of executing high jumps. Robots of the mass scale of ) cam mechanisms for jumping and generally jump higher than the lighter jumpers. The heavier robots (over use more sophisticated mechanical solutions such as electro-mechanical clutches, but the jumping height does not necessarily compensate for the extra weight of these mechanisms. 2. Bio-inspired Design Inspired by the desert-locust, the focus is on the main bio-mechanical principles of the locust’s jump rather than attempting to produce an exact imitation of its body and morphology in order to create a simple, yet effective, jumping mechanism. In this study, the demonstrated mechanism is closer to nature’s design and has advantages over other miniature jumping robots. The suggested artificial mechanism, differs from previous designs in its energy storage and release subsystems. The jumping robot is designated TAUB (Tel-Aviv University and Ort-Braude College). The jumping mechanism utilizes a pair of legs, each with two equal segments, inspired by the hind legs of the locust. Torsion springs, as the elastic elements, are located in the joints between the two segments of each leg and connect them. The ‘muscle’ of the mechanism is an electric motor connected directly to the lower part of the leg by wire. The principle is to use a miniature motor, and through a wire to transfer the torque to deform the springs located away from the motor. Finally, the legs of the jumping mechanism are connected to the robot’s body through a revolute joint, similar to the connection of the locust’s legs to its body. The elements of the conceptual design, and the desert locust are shown in Fig.2. Body (motor, battery etc.) Torsion springs

Jumping legs

Jumping legs

Wire (a)

(b)

Fig. 1. (a) Conceptual locust-inspired design of the jumping robot; (b) Desert-locust.

The jump sequence has two phases. In the first phase, the loading phase, the rotation of the motor causes the tendon-like wire to coil on the motor’s axis and pull the tip of the legs towards the body. As the leg approaches the body, the torsion springs deform, storing the required elastic energy for the jump. When the tip of the leg is close to the body, the springs are locked at their full capacity and the robot is in the initial position for the jump. In the second phase, the releasing phase, the motor switches its direction of rotation, uncoiling the wire. Once the wire is completely uncoiled, the legs are released and the springs speedily release their energy. The legs' structure then converts the elastic energy into kinetic energy, resulting in the required leap speed that propels the robot into the air, similar to the locust’s and other legged-insects' jump.

4

Uri Ben Hanan et al. / Procedia Manufacturing 21 (2018) 243–250 Uri Ben Hanan / Procedia Manufacturing 00 (2017) 000–000

247

Most of the robotic designs, rely on cam mechanisms for storing and releasing the jumping energy. TAUB's suggested mechanism has an advantage in force/torque enhancement over these existing mechanisms. The transmission ratio between the cam output torque and the spring's force/torque is approximately 1/5 to 1/10, resulting in maximal torque enhancement of 5 to 10 times the motor's output torque. In TAUB's mechanism (see Fig.2(a)), the torque enhancement is caused by a natural lever ratio obtained by the length of the legs and the motor's axle radius. The practical transmission ratio is around 1/50 thus making it more effective than cam mechanisms for torque enhancement. Another advantage of TAUB's mechanism is that the springs have a relatively large angle of , similar to the locust [10]. Compared to other designs, such as the EPFL jumper [16] deformation, about and the MSU jumper [18], TAUB's mechanical advantage is far superior. This, in turn, allows lowering the with springs' stiffness, decreasing the load on the joints and legs, and storing more energy with no weight increase. In order to fully realize the suggested concept for a jumper, a prototype, shown in Fig. 3 was designed. The prototype uses a geared miniature motor, a small Li-Po battery, a tiny microcontroller, and jumping legs. The leg structure is composed of carbon rods, and is designed to include maximal amount of torsion springs without yielding geared servo electric motor (HS-35HD manufactured by HITEC), with under their load. The prototype uses a , the highest among all the other evaluated motors. In the prototype, a stall-torque-to-mass ratio of the equivalent spring stiffness is adjusted by using any number of springs from two to eight. After optimization of diameter and length carbon rods were selected for the jumping legs, the strength-to-mass ratio, . combined with eight light steel torsion springs, each with a spring coefficient of

Torsion springs

Controller

Motor

Jumping legs

Battery

Wire (b)

(a)

Fig. 2. (a) Prototype in released position; (b) Prototype in loaded position.

A microcontroller (Attiny 85 BGA) controls the prototype. The control unit is designed for minimal weight, with only the electronics necessary to operate the robot. The geared motor is supplied with power by a small and Li-Po battery with capacity (manufactured by Full River) weighing . The lightweight where is the elastic approximate elastic energy used for loading the springs is evaluated using energy, yielding roughly . In order to evaluate the battery life of TAUB, the constructed prototype (see Fig.3) was tested to perform repeated loadings, using a fully charged Li-Po battery. It delivered 123 repeated loadings before the battery was exhausted. 3

Height [m]

2.5 2 1.5 1 0.5 0 0

(a)

0.5

Distance [m]

1

1.5

(b)

Fig. 3. (a) Jump ballistics filmed in 240 fps; (b) The trajectories of 10 jumps (grey points) and the average trajectory (solid red line).

Uri Ben Hanan et al. / Procedia Manufacturing 21 (2018) 243–250 Uri Ben Hanan / Procedia Manufacturing 00 (2017) 000–000

248

5

The overall weight of the prototype is . The design and construction of the prototype aimed at reducing the mass; however, mechanical strength is a consideration that limits further mass reduction. 3. Jumping Experiments and Results The behaviour of one of the highest jumps of the prototype was captured using slow motion filming with 240 fps camera (Fujifilm HS- 10) is shown Fig.4(a). Table 2. Jumping experimental results of 12 jumps Number of jump

Height (m)

Distance (m)

Leap angle (deg)

Angular velocity (rad/sec)

1 2 3 4 5 6 7 8 9 10 11 12 Mean STD

3.67 3.24 3.63 3.13 3.15 3.53 3.36 3.41 3.19 3.20 3.66 3.13 3.35 0.21

1.23 1.32 1.12 1.07 0.94 0.94 0.46 1.88 0.70 1.18 2.41 3.2 1.37 0.77

84 82.9 82.9 81.7 81.5 83.8 83.8 77.2 85.5 83 78.8 73.1 81.5 3.5

41.16 38.08 38.08 42.99 39.05 41.16 36.02 42.31 39.19 40.38 27.19 33.40 39.84 2.15

Table 3. Comparison between recent miniature jumping robots Robot (year)

Mass (gr)

Jumping height (cm)

Height to mass ratio (cm/gr)

Jumping distance (cm)

[23], 2006 [24], 2010

54.1

57 20

0.37

[25], 2014 [19], 2012 [16], 2008 [17], 2009 [14], 2012

25 1.104 7 9.8 7

143 64 138 76 71

5.72 57.97 19.71 7.75 10.14

59

[26], 2011

16.5

12

[22], 2014

67.5

[20], 2012 [27], 2010 [4], 2007

Onboard energy & control

Jumping period (sec)

Actuation

Running speed 5 cm/s

1 motor, self-righting mechanism, spring load and release mechanism

21 3.5 3.5

100

Yes No Yes Yes No

073

30.2

Yes

3

100

1.48

over 800

Yes

18 700

8 15 160

0.83 0.23

95 220

Yes No Yes

0.1 0.32

[18], 2013

23.5

90

3.83

Yes

7

TAUB, 2015

23

335

14.57

Yes

20

3 SMA 1 motor & cam mechanism 1 motor & cam mechanism 1 motor & cam mechanism, linear spring 2 motors, cam mechanism selfrighting, steering 1 motor, clutch, carbon fiber spring, pivoting wing, Chemical energy 2 motors 1 motor, linear spring, release using clutch 1 motor, pully, rotation link and torsional springs 1 motor & latch mechanism, torsional spring

Yes Yes

137

6

Uri Ben Hanan et al. / Procedia Manufacturing 21 (2018) 243–250 Uri Ben Hanan / Procedia Manufacturing 00 (2017) 000–000

249

The recorded jumping height is and the jumping distance is as well, with flight duration. In addition, an angular velocity was observed during the ballistic in-flight duration, resulting in uncontrolled landing of the robot. Twelve highest jumps were selected for statistical evaluation. For this purpose, detailed analysis was conducted using the video analysis software Tracker. Table 2 presents the results and statistics of the twelve jumps. with a STD of , the mean distance is with a STD of , the mean The mean height is with a STD of and the mean angular velocity is with a STD of . leap angle is . The highest jump achieved was Fifty five jumping tests were conducted to fully characterize TAUB's jumping mechanism, of which, the ten closest trajectories were selected, not necessarily optimal in height nor distance. The selected jumps (see Fig.4(b)) were digitized using 50 points to represent each jump (marked by grey points in Fig.4(b)). A parabola describing a perfect ballistic trajectory was fitted by least-mean square analysis to the selected trajectories (red line in Fig.4(b)). . The height of the fitted ballistic trajectory is Table 3 shows compares TAUB's jumping mechanism to some of the state-of-the-art existing jumpers, to emphasize the contribution of this design. 4. Summary and Discussion This paper presents the design and performance of TAUB, a jumping robot inspired by the desert-locust. TAUB's jumping mechanism mimics the legs of the locust and uses a similar pair of legs structure in order to jump. TAUB's artificial jumping mechanism however, is not a copy of the locust but uses only its main features for the jump, such as storing energy in the joints of the legs and producing a 'kick' motion by rapidly releasing this energy. Fig.5 presents TAUB's CAD design next to the prototype. The suggested solution combines locust-like legs design with a miniature motor and a wire. The negligible-mass wire, serves as the torque/force enhancer, replacing any other force multipliers that usually add extra weight, such as a cam mechanism.

(a)

(b)

Fig. 4. (a) CAD model of TAUB; (b) TAUB prototype (loaded position)

The prototype uses a state-of-the-art small-geared motor, combined with small-scale power source and tiny microcontroller. The energy to mass ratio is optimized by adding springs up to point of nearly-yielding the carbon legs, with overall mass reduction to keep the system's weight as light as possible. The overall mass of the prototype . The prototype was tested and achieving an average jumping height of , and a horizontal distance of is . The maximal jumping height recorded was and the highest jumping distance was . TAUB’s jumping mechanism is the foremost among all its predecessors of the same size-scale. TAUB reaches more than twice the height of its predecessors (see table 3). The most promising cam based robot is the EPFL jumper and its body [16] that achieved a jumping height to body length ratio of 27 (its maximal jumping height is ). TAUB's energy storage and release mechanism, however, enabled better performance of 31 body length is lengths (legs closed, ready to jump, see Fig.3(b)). The main advantage of TAUB's design, compared to other robots is its energy storage capacity with respect to its weight, which yields higher jumps. The wire based torque enhancement is large compared with cam mechanisms by a factor of at least five to ten. Such enhancement is possible because the springs are located away from the body, , and obtains which is similar to the configuration in a locust. This enables a large deflection angle of about higher transmission ratio between the motor's shaft and the springs. Thus, more springs are added to the joint

250

Uri Ben Hanan et al. / Procedia Manufacturing 21 (2018) 243–250 Uri Ben Hanan / Procedia Manufacturing 00 (2017) 000–000

7

efficiently using the motor's torque. Finally, it can be concluded that this paper serves as an example for successful incorporation of bio-inspired design principles, into a robot's design process. References [1] De Cubber G, Serrano D, Berns K, Chintamani K, Sabino R, Ourevitch S, Doroftei D, Armbrust C, Flamma T, Baudoin Y. Search and rescue robots developed by the european icarus project. In7th Int. Workshop on Robotics for Risky Environments 2013 Oct 1. [2] Seeni A, Schäfer B, Hirzinger G. Robot mobility systems for planetary surface exploration–state-of-the-art and future outlook: a literature survey. InTech; 2010. [3] Kaspari M, Weiser MD. The size–grain hypothesis and interspecific scaling in ants. Functional Ecology. 1999 Aug 1;13(4):530-8. [4] Armour R, Paskins K, Bowyer A, Vincent J, Megill W. Jumping robots: a biomimetic solution to locomotion across rough terrain. Bioinspiration & biomimetics. 2007 Jun 22;2(3):S65. [5] Dubowsky S, Kesner S, Plante JS, Boston P. Hopping mobility concept for search and rescue robots. Industrial Robot: An International Journal. 2008 May 2;35(3):238-45. [6] Hougen DF, Benjaafar S, Bonney JC, Budenske JR, Dvorak M, Gini M, French H, Krantz DG, Li PY, Malver F, Nelson B. A miniature robotic system for reconnaissance and surveillance. InRobotics and Automation, 2000. Proceedings. ICRA'00. IEEE International Conference on 2000 (Vol. 1, pp. 501-507). IEEE. [7] Bennet-Clark HC, Lucey EC. The jump of the flea: a study of the energetics and a model of the mechanism. Journal of Experimental Biology. 1967 Aug 1;47(1):59-76. [8] Parry DA, Brown RH. The jumping mechanism of salticid spiders. Journal of Experimental Biology. 1959 Dec 1;36(4):654-64. [9] Burrows M. Jumping performance of froghopper insects. Journal of Experimental Biology. 2006 Dec 1;209(23):4607-21. [10] Heitler, W., 2007. How Grasshoppers jump, from https://www.st-andrews.ac.uk/~wjh/jumping/ [11] Queathem EJ, Full RJ. Variation in jump force production within an instar of the grasshopper Schistocerca americana. Journal of Zoology. 1995 Apr 1;235(4):605-20. [12] Evans ME. The jump of the click beetle (Coleoptera, Elateridae)—a preliminary study. Journal of zoology. 1972 Jul 1;167(3):319-36. [13] Bennet-Clark HC. The energetics of the jump of the locust Schistocerca gregaria. Journal of Experimental Biology. 1975 Aug 1;63(1):53-83. [14] Nguyen QV, Park HC. Design and demonstration of a locust-like jumping mechanism for small-scale robots. Journal of Bionic Engineering. 2012 Sep 1;9(3):271-81. [15] Kai Z, Junmao Y, Diansheng C, Tianmiao W. Design and kinematics simulation for bionic crank-slider mechanism of jumping robot. InRobotics and Biomimetics (ROBIO), 2012 IEEE international conference on 2012 Dec 11 (pp. 796-801). IEEE. [16] Kovac M, Fuchs M, Guignard A, Zufferey JC, Floreano D. A miniature 7g jumping robot. InRobotics and Automation, 2008. ICRA 2008. IEEE International Conference on 2008 May 19 (pp. 373-378). IEEE. [17] Kovač M, Schlegel M, Zufferey JC, Floreano D. A miniature jumping robot with self-recovery capabilities. InIntelligent Robots and Systems, 2009. IROS 2009. IEEE/RSJ International Conference on 2009 Oct 10 (pp. 583-588). IEEE. [18] Zhao J, Xu J, Gao B, Xi N, Cintrón FJ, Mutka MW, Xiao L. MSU jumper: A single-motor-actuated miniature steerable jumping robot. IEEE Transactions on Robotics. 2013 Jun;29(3):602-14. [19] Noh M, Kim SW, An S, Koh JS, Cho KJ. Flea-inspired catapult mechanism for miniature jumping robots. IEEE Transactions on Robotics. 2012 Oct;28(5):1007-18. [20] Churaman WA, Currano LJ, Morris CJ, Rajkowski JE, Bergbreiter S. The first launch of an autonomous thrust-driven microrobot using nanoporous energetic silicon. Journal of Microelectromechanical Systems. 2012 Feb;21(1):198-205. [21] Ackerman E. Boston dynamics sand flea robot demonstrates astonishing jumping skills. IEEE Spectrum Robotics Blog. 2012 Mar;2(1). [22] Desbiens AL, Pope MT, Christensen DL, Hawkes EW, Cutkosky MR. Design principles for efficient, repeated jumpgliding. Bioinspiration & biomimetics. 2014 May 22;9(2):025009. [23] Stoeter SA, Papanikolopoulos N. Kinematic motion model for jumping scout robots. IEEE transactions on robotics. 2006 Apr;22(2):397402. [24] Zhao J, Xi N, Gao B, Mutka MW, Xiao L. Design and testing of a controllable miniature jumping robot. InIntelligent Robots and Systems (IROS), 2010 IEEE/RSJ International Conference on 2010 Oct 18 (pp. 3346-3351). IEEE. [25] Zhao J, Yan W, Xi N, Mutka MW, Xiao L. A miniature 25 grams running and jumping robot. InRobotics and Automation (ICRA), 2014 IEEE International Conference on 2014 May 31 (pp. 5115-5120). IEEE. [26] Kovač M, Fauria O, Zufferey JC, Floreano D. The EPFL jumpglider: A hybrid jumping and gliding robot with rigid or folding wings. InRobotics and Biomimetics (ROBIO), 2011 IEEE international conference on 2011 Dec 7 (pp. 1503-1508). IEEE. [27] Yamada A, Mameda H, Mochiyama H, Fujimoto H. A compact jumping robot utilizing snap-through buckling with bend and twist. InIntelligent Robots and Systems (IROS), 2010 IEEE/RSJ International Conference on 2010 Oct 18 (pp. 389-394). IEEE.