Stability Improvement Using Soft Sole on Humanoid Robot

Masanao Koeda1, Tsuneo Yoshikawa1, and Keita Nakagawa1

1Ritsumeikan University, College of Information Science and Engineering,Department of Human and Computer Intelligence

Noji Higashi 1-1-1 Kusatsu, 525-8577 Shiga, Japan.email: {koeda, yoshikawa, nakagawa}@robot.ci.ritsumei.ac.jp

Abstract— This paper shows the effect of a stabilityimprovement of humanoid robot with a soft sole of feetwhich is aimed at walking on an uneven terrain. Gener-ally, it was thought that the soft sole absorbs shocks butdecrease stability at the same time. We conducted pre-liminary experiments of shifting center of gravity, step-ping motions, and walking on an even/uneven terrainwith and without soft sole of the feet. The experimentalresults show that the soft sole increases the stability incertain conditions.

Keywords— Humanoid Robot, Soft Sole, Stability, Un-even Terrain

I. Introduction

RECENTLY , various humanoid robots have beendeveloped[1], [2]. For humanoid robots, walking

skill is one of the most important and basically func-tions. However, at this time, it is difficult for humanoidrobots to walk adaptively on the floor conditions suchas a deformable floor like a grass plot or a sandy place,or an uneven terrain like a gravel road. Researches towalk on an uneven terrain by a humanoid robot can beclassified into category; Active and Passive. The formeraims for recognition of the condition of the floor usinga sensor such as a stereo camera, or an ultrasonic sen-sor [3], [4], and generation of gait pattern depending onthe floor condition [5], [6], [7]. Yamaguchi et al.[8] de-veloped mechanical foot with shock absorbing materialand investigated its stability.

Generally, it is known empirically that the soft soleabsorbs shocks but decrease stability at the same time.We investigated the stability of a humanoid robot witha soft sole at the feet’s bottom. Experiments were con-ducted in following 3 motions; shifting the center ofgravity, stepping, and walking on an even/uneven ter-rain, and stability margin was compared each of themwith the case of not using any rubber foam. Finally,we show that the stability was improved in a certaincondition contrary to our expectations.

II. System Overview

Our system consists of a humanoid robot HOAP-2(Fig.1) and a host PC. The system overview is shownin Fig.2.

A. HOAP-2

HOAP-2 is a commercial humanoid robot which is de-veloped by FUJITSU AUTOMATION Ltd. Its heightis 50[cm] and weight is 7[kg] approximately. The sizeof foot is 98[mm] by 63[mm]. Detailed specification isshown in TABLE I. In each foot, 4 force sensors aremounted, and 3 axis acceleration/angular sensors are

Fig. 1. Humanoid robot HOAP-2

Host PCRT-Linux

HOAP-2

USB

DC 24V

Fig. 2. System configuration

equipped in its body. Fig.4 illustrates the size of thefeet and the position of the force sensors. CH.0, 1, 2and 3 show the position of the force sensors.

B. Host PC

The robot is controlled in 1[ms] through a host PCwhich is running RT-Linux OS. They are connected byUSB interface. The PC sends control commands to therobot, and the robot transmits acquired data from sev-eral sensors. A wireless network based on 802.11b canbe used for communication between the PC and therobot. However, we selected the wire communicationbecause the moving range of the robot is not so large inthis study.

Fig. 3. Attached 10[mm] rubber foam

Fig. 4. Foot size and sensors position

C. Soft Sole

6 kinds of rubber foams of 5, 10, 15, 20, 25, and30[mm] thickness were used as the soft material. TheYoung’s modulus of the rubber foam is 45.3[Pa]. Thefoams are cut in the same size as the robot’s foot andattached to the sole by a thin two-sided tape. Fig.3shows the feet which is attached with 10[mm] rubberfoam. When no rubber foam is attached, it is called“0[mm] thickness” afterward.

III. Evaluating Method

A. Stability Margin

To calculate stability of the robot quantitatively,”Stability margin” is defined as the following equationfor evaluation criteria.

(Stability Margin) = min {di | i = 1, 2, 3, 4} (1)

where di means the distance between ZMP(Zero Mo-ment Point)[9] to each edge of the sole. In Fig.5, d3

becomes stability margin because of the shortest dis-tance from ZMP to the right edge. It can be said thatthe stability is improved when the stability margin be-comes large.

ZMP

d1

d3d2

d4

Sole

Fig. 5. Stability margin

TABLE ISpecifications of HOAP-2

Height 50[cm]Weight 7[kg]DOF 6 in Leg × 2

4 in Arm × 21 in West × 11 in Hand × 22 in Neck × 1(Total 25DOF)

Sensors Joint Angle SensorResolution: 0.01[deg/pulse]

3 Axis AccelerometerMeasuring Range: ±2[G]Resolution: 0.005[G]

3 Axis GyrometerMeasuring Range: ±60[deg/s]Resolution: 0.25[deg/s]

Foot Load Sensor4[ch/foot]

PC OS: RT-LinuxCPU: Pentium4 2[GHz]Memory: 1[GB]Sampling Rate: 1[ms]

IV. Experiments and Results

We conducted the following 4 set of experiments;shifting center of gravity, stepping motions, and walk-ing on an even/uneven terrain with and without thesoft sole of the feet. To compare the experimental re-sults easily, all motions were generated by a completeopen loop without feedback from any sensors.

A. Shifting Center of Gravity

A.1 Right and Left

To move the center of gravity to the right and to theleft, the leg was tilted slowly by 10[deg] without liftingthe foot while the body was horizontal. This motion wasoperated for 20[sec]. Fig.7 shows the sequential imagesof this motion. This motion was operated 3 times in the

10

15

20

25

30

35

0 1 2 3 4 5 6 7 8 9 10

Sta

bilit

y M

argi

n [m

m]

Time [sec]

0mm sole

5mm sole10mm sole

(a) Right foot

10

15

20

25

30

35

10 11 12 13 14 15 16 17 18 19 20

Sta

biity

Mar

gin

[mm

]

Time [sec]

0mm sole

5mm sole

10mm sole

(b) Left foot

Fig. 6. Change of Stability Margin while shifting COG

(a) Initial pose (0[s]) (b) Tilt at 10[deg] tothe right (5[s])

(c) Center (10[s]) (d) Tilt at 10[deg] tothe left (15[s])

(e) End (20[s])

Fig. 7. Shifting COG to the right and left

(a) Initial pose (0[s]) (b) Tilt at 5[deg] toforward (5[s])

(c) Center (10[s]) (d) Tilt at 5[deg] tobackward (15[s])

(e) End (20[s])

Fig. 8. Shifting COG to backward and forward

conditions that 7 kinds of rubber foam (from 0[mm] to30[mm] thickness) was attached. Fig.6 shows temporalchanges of the stability margin while 3 kinds of sole (0,5, and 10[mm]). The horizontal axis shows time andthe vertical axis shows the stability margin. In consid-eration of the measuring range of the force sensor, ZMPof the right foot in 0-10[sec] and ZMP of the left foot in10-20[sec] were used to calculate the stability margins.The stability margins have decreased along with theshifting center of gravity. Moreover, the lowest valueof the stability margin increases by attaching rubberfoams. In the following experiments, the lowest valuesof the stability margin are compared. Fig.9-(a) showsthe average of minimum stability margin in 7 kinds ofsoles. The stability margin became the maximum at20[mm], and it tended to decrease in thicker rubber.

A.2 Backward and Forward

To move the center of gravity to backward and for-ward, the leg was tilted slowly by 5[deg] without liftingthe foot while the body was horizontal. This motionwas operated for 20[sec], and Fig.8 shows the sequen-

tial images of this motion. Fig.9-(b) shows the averageof minimum stability margin in 7 soles. It shows thatthe rubber does not influence the stability margin somuch in this motion.

B. Stepping Motion

Next, the stability margins while the spot steppingmotion were compared. Experimental conditions werethat one step took 8[sec] and each foot was lifted by20[mm]. Fig.10 shows the sequential images of this mo-tion. The averages of minimum stability margins whilethe motion was measured for 7 kinds of sole in 5 timesrespectively. The results are shown in Fig.11. To calcu-late the stability margin in the measuring range of theforce sensor, the ZMP of the landing foot was calculatedin 0-2[sec], 6-8[sec], 8-10[sec], and 14-16[sec]. The re-sults shows that the stability had decreased when over25[mm] thickness rubber was attached, and the sametendency appeared as the previous experiments.

0

5

10

15

20

25

30

0 5 10 15 20 25 30

Min

imum

Sta

bilit

y M

argi

n [m

m]

Thickness of Sole [mm]

(a) Right and left

0

5

10

15

20

25

30

0 5 10 15 20 25 30

Min

umum

Sta

bilit

y M

argi

n [m

m]

Thickness of Sole [mm]

(b) Backward and forward

Fig. 9. Stability margin while shifting COG motion

(a) Initial position(0[s])

(b) Shift COG toright (2[s])

(c) Lift left footvertically (4[s])

(d) Lower left leg(6[s])

(e) Shift COG tocenter (8[s])

(f) Shift COG to left(10[s])

(g) Lift right footvertically (12[s])

(h) Lower right leg(14[s])

(i) Shift COG tocenter (16[s])

Fig. 10. Stepping motion

(a) Initial position(0[s])

(b) 1st step (4[s]) (c) 2nd step (10[s]) (d) 3rd step (16[s]) (e) 4th step (22[s])

(f) 5th step (28[s]) (g) 6th step (34[s]) (h) 7th step (40[s]) (i) 8th step (46[s]) (j) End (48[s])

Fig. 13. Even terrain walking with 10[mm] sole

C. Walking Motion

C.1 On Even Terrain

Experimental conditions were that one step took6[sec], each foot was lifted by 20[mm], the length ofstride was 30[mm], and the robot took 8 steps forward.

The experiments were conducted 7 times in each solerespectively, and the stability margin and the numberof non-fall were compared. Fig.13 shows the sequentialimages of this motion. Fig.12 shows the average of min-imum stability margin of each experiment. Same as in

0

5

10

15

20

25

30

0 5 10 15 20 25 30

Min

imum

Sta

bilit

y M

argi

n [m

m]

Thickness of Sole [mm]

Fig. 11. Stability margin on stepping motion

0

5

10

15

20

25

30

0 5 10 15 20 25 30

Min

imum

Sta

bilit

y M

argi

n [m

m]

Thickness of Sole [mm]

Fig. 12. Stability margin on even terrain walking

the case of the stepping motion, the ZMP of the landingfoot was used. In whole of the conditions, the stabil-ity margin became small since the walking motion wasunstable itself, and it shows a similar tendency as theprevious experiments. Fig.17-(a) shows the number ofnon-fall while the robot takes 8 steps in 7 trials. Therobot did not fall over when the rubber foam below15[mm] thickness was attached on the sole. However,when the rubber was over 20[mm] thickness, the robottended to become unstable and fall over.

C.2 On Uneven Terrain

Finally, we confirm the effect of the soft sole whilewalking on an uneven terrain. The motion is the sameas the previous one. Fig. 14 shows the condition of theconstructed uneven terrain for this experiment. Hemi-spheric convex objects of 5[mm] in diameter are ar-ranged by every 80[mm] on a woody board. The charac-ter S in Fig.14 indicates the starting position and the 7arrows which were drawn every 15[deg] mean the start-ing pose of the robot. Fig. 15 and 16 are the experi-mental snapshots when the robot which has 0 or 10[mm]rubber sole is walking in the direction of the red arrowin Fig.14. The numbers of non-fall in 7 trials are indi-cated in Fig.17-(b).

The experimental results shows that the excessivethickness of the rubber makes stability decrease, andthe thickness about 10[mm] is suitable for walking onan uneven terrain.

650

400

18

9

5

[mm]

15[deg]

80 80 80 80 80 80 80

8080

8080

S

Fig. 14. Condition of uneven floor

V. Conclusion

In this paper, we described the stability improvementof a humanoid robot which was attached soft materi-als on its soles. We conducted experiments of shiftingcenter of gravity, stepping motions, and walking on aneven/uneven terrain with and without soft sole of thefeet. As a result, comparing the soft sole with the hardsole in the point of view of the stability margin, softsoles increase stability in certain conditions. Proposedmethod is quite simple remodeling to improve the sta-bility, and it is effective for an irregular terrain.

In the future, we will analyze this effect mathiemat-ically, and investigate the optimal thickness, hardnessand shape of the sole, and we will also establish a suit-able control method.

References

[1] Y. Sakagami, R. Watanabe, C. Aoyama, S. Matsunaga, N.Higaki, and K. Fujimura, “The intelligent ASIMO: Sys-tem overview and integration”, In Proceedings of 2002IEEE/RSJ International Conference on Intelligent Robotsand Systems (IROS), pp. 2478-2483, 2002.

[2] K.Kaneko, F.Kanehiro, S.Kajita, H.Hirukawa, T.Kawasaki,M.Hirata, K.Akachi, and T.Isozumi, “Humanoid RobotHRP-2”, In Proceedings of 2004 IEEE International Confer-ence on Robotics and Automation (ICRA), pp. 1083-1090,2004.

[3] T. Yoshimi, Y. Kawai, Y. Fukase, H. Araki, and F. Tomita,“Measurement of Ground Surface Displacement using StereoVision and Mechanical Sensors on Humanoid Robots”,In Proceedings of 2003 IEEE International Conference onMultisensor Fusion and Integration for Intelligent Systems(MFI), pp.125-130, 2003.

[4] S. Kajita, and K. Tani, “Adaptive gait control of a bipedrobot based on realtime sensing of the ground”, In Proceed-ings of 1996 IEEE International Conference on Robotics andAutomation (ICRA), pp.570-577, 1996.

[5] R. Ozawa, Y. Takaoka, Y. Kida, K. Nishiwaki, J. Chestnutt,J. Kuffner, S. Kagami, H. Mizoguchi, and H. Inoue, “Us-ing Visual Odometry to Create 3D Maps for Online Foot-step Planning”, In Proceedings of 2005 IEEE InternationalConference on Systems, Man and Cybernetics (SMC), pp.2643-2648, 2005.

[6] H. Takemura, A. Matsuyama, J. Ueda, Y. Matsumoto, H.Mizoguchi, and T. Ogasawara, “Momentum Compensationfor the Dynamic Walk of Humanoids based on the OptimalPelvic Rotation”, In Proceedings of 2005 International Con-ference on Climbing and Walking Robots and the Support

(a) Initial position(0[s])

(b) 1st step (4[s]) (c) 2nd step (10[s]) (d) 3rd step (16[s]) (e) 4th step (22[s])

(f) 5th step (28[s]) (g) 6th step (34[s]) (h) Fall (38[s])

Fig. 15. Uneven terrain walking with 0[mm] sole

(a) Initial position(0[s])

(b) 1st step (4[s]) (c) 2nd step (10[s]) (d) 3rd step (16[s]) (e) 4th step (22[s])

(f) 5th step (28[s]) (g) 6th step (34[s]) (h) 7th step (40[s]) (i) 8th step (46[s]) (j) End (48[s])

Fig. 16. Uneven terrain walking with 10[mm] sole

0

1

2

3

4

5

6

7

0 5 10 15 20 25 30

Num

ber o

f Non

-Fal

l

Thickness of Sole [mm]

(a) Even terrain walking

0

1

2

3

4

5

6

7

0 5 10 15 20 25 30

Num

ber o

f Non

-Fal

l

Thickness of Sole [mm]

(b) Uneven terrain walking

Fig. 17. Number of non-fall while walking

Technologies for Mobile Machines (CLAWAR), pp.485-492,2005.

[7] S. Kajita, K. Kaneko, K. Harada, F. Kanehiro, K. Fujiwara,and H. Hirukawa, “Biped Walking On a Low Friction Floor”,In Proceedings of 2004 IEEE/RSJ International Conferenceon Intelligent Robots and Systems (IROS), pp. 3546-3552,2004.

[8] J. Yamaguchi, A. Takanishi, and I. Kato, “Experimental De-velopment of a Foot Mechanism with Shock Absorbing Mate-

rial for Acquisition of Landing Surface Position Informationand Stabilization of Dynamic Biped Walking”, In Proceed-ings of 1995 IEEE International Conference on Robotics andAutomation (ICRA), pp.2892-2899, 1995.

[9] P. Sardain and G. Bessonnet, “Forces Acting on a BipedRobot, Center of Pressure – Zero Moment Point”, IEEETransactions of Systems, Man, and Cybernetics, Part A. Vol.34, No. 5, pp. 630-637, 2004.