The Bearing Capacity and the Rational Loading Mode of Pigeon during
Tingting LIU1,2,a, Lei CAI1, Hao WANG*,1,b, Zhendong DAI1, Wenbo WANG1
1Institute of Bio-inspired Structure and Surface Engineering, Nanjing University of Aeronautics and
Astronautics, Nanjing 210016, china
2college of automation engineering, Nanjing University of Aeronautics And Astronautics, Nanjing
aemail: email@example.com, bemail: firstname.lastname@example.org
*corresponding author: Hao WANG
Keywords: pigeon-robot; bearing capacity; loading mode; pigeon (Columba livia)
A bio-robot treats the living animal itself as the robot-moving carrier, which possesses good
mobility and adaptability even in complicated non-structured environments. One of the flyable
bio-robots（pigeon-robots）have been studied for several years. However, the arrangement of
equipment on board effectively to minimize the impact on pigeons’ natural flight has not been
considered. In this study, we estimated the bearing capacity of a pigeon during takeoff by increasing
the weight gradually. Different loading modes have also been compared to find a rational way to
reduce the influence of the loading to flyability. The results are expected to be helpful to design
As a common bird in our daily lives, pigeon’s good orientation discrimination, enduring flight
and load carrying flyability have impressed many scholars. The pigeon-robot  has a high practical
value both in military and civilian application. For example, it can be used in the tasks of
investigation, radar interference, the wild tracing and resources exploration. To successfully
perform these tasks, pigeon would be equipped with some special devices [2-4], which propose a
very high demand on the load-bearing capacity of pigeon.
So far, few studies concerning animal’s bearing ability during flight have been conducted
specifically and systematically. Most scholars focused on the animal’s mechanical energy
consumption or some birds’ flying kinematics and aerodynamics under conditions of loading [5-9].
Animal’s maximum bearing capacity may involve many factors such as temperature, altitude, wind
characters and flying space. Different species of animals may have different load carrying abilities.
Patsy M. Hughes et al.  studied the long-eared bat’s flight performance with the addition of
artificial wing loading. With the increase in wing loading, the bat’s flying speed decreased . Six
ruby-throated hummingbirds’ hovering performance under conditions of maximum load had been
studied by Peng Chai who described a way of measuring the maximal load-lifting capacity of
hummingbirds . Robert L. Nudds et al. revealed that the costs of carrying additional mass of
finches during routine short flights were behavioral and not energetic . James H. Marden showed
that the bird’s bearing capacity during flying was associated with many aspects of the animal itself
and he compared the impact on the production of the muscle mass-specific lift which was related to
Applied Mechanics and Materials Vol. 461 (2014) pp 122-127
Online available since 2013/Nov/21 at www.scientific.net
© (2014) Trans Tech Publications, Switzerland
All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of TTP,
www.ttp.net. (ID: 126.96.36.199-24/11/13,14:26:19)
the bearing capacity . Meanwhile the loading modes may also have some influence on the
research results. Peter G. Tickle et al. investigated the effect of load placement on the energy
consumption of barnacle goose . Therefore, the arrangement of equipment on board effectively to
minimize the impact on pigeons’ natural behavior requires investigation.
In this study we measured the maximum load bearing capacity of pigeon during takeoff and
compared the effects of two different loading modes. This experiment is expected to play an
important referential role in research and design of pigeon-robots.
Materials and Methods
Animals were healthy adult pigeons (Columba livia) with average weight of 484.8±17.6 g. Five
pigeons with intact flight feathers marked 1 to 5 were served as subjects and were housed in a
dovecote. They were supplied with enough food and water and were released every day. The
experimental procedures were in accordance with the animal protection department’s guidelines.
Experimental Arena and Experimental Tools
Loading trials were conducted in a cage with dimensions of 2 m×1.2 m×1.8 m (L×W×H).
Canvas with Julesz pattern was attached to the four walls of the cage in order to prevent outside
visual interference and to help pigeons to recognize and remember experimental environment. The
pigeon’s minimum resolution angle was about 25'～39'  and the length of the cage was 2 m. The
minimum resolution distance was 0.024 cm～0.029 cm and choosing 5 cm as the random grid side
length was enough for the pigeon to discriminate. Food containers and water box were placed on
another wall of the cage. A 1.4 m high habitat rod was hung near the food containers and water box.
The rod was rough enough to ensure that the pigeons could stand on it stably. A strap penetrating
from both sides of the wing bottom of the pigeon was made and a bag filled with weights (per unit
weight 5 g) was connected to the strap.
Experimental Specifications and Experimental Procedures
Certain criterions were defined for a scientific analysis. Here we defined successful takeoff as
those flights that pigeons could lift the loads to the rod within 5 min as soon as they got into the
experimental arena. The whole experimental process consisted of three stages. The first stage was
training stage including getting familiar with the environment and flying with load training. The
second stage was the maximum load measuring stage and the last stage was statistical analysis stage.
The training stage was necessary for the first trial. Let five starving pigeons be in the arena at least
one day to know the condition well. Through the pre-experimental experience we knew that pigeons
which had not been trained to fly with load seldom had the awareness of flying upward with load on
their bodies. The weight used in this stage was only 72 g. Trained pigeons would be used in the
second stage in which they need not to be starved. Here we defined the basic weight as 72 g and per
unit weight as 5 g. During the trail we got the maximum load by increasing the weight gradually
until the pigeon was not able to take off successfully. The 5 pigeons’ trials were conducted for three
rounds in turns. The pigeon with the maximum load must take off successfully at least twice in the
three round trials. Between each round a pigeon was allowed to take food for 30 s on the rod and
taken away to rest in a room without interference. When the pigeons were very familiar with the
arena, the training stage could be omitted. The last analysis stage using the software SPSS 10.0 and
GraphPad Prism V5.01 to process the result data. As to the loading mode experiment, we chose the
load placements below the xiphoid and on the back of pigeon respectively. A view of load
placement on a pigeon is given in Fig. 1.
Applied Mechanics and Materials Vol. 461 123
The pigeon was weighted before each trial. The maximum bearing capacity was described as the
ratio of the maximum load to pigeon’s weight. The loading mode experiment was not conducted on
the same day, so the body mass of a pigeon was different. As this experiment aimed at obtaining the
ratio value, the impact of different body mass for the same pigeon could be neglected. Table 1
showed the results of the two different loading modes. The maximum load below the xiphoid could
reach to about 25%～27% of the body mass. The overall trend of bearing capacity was that pigeon
could lift more loads below the xiphoid than on the back.
Table 1 Outcome of the maximum load below the xiphoid and on the back. X means the mode
below xiphoid and b means the mode on the back.
Fig. 2 showed the scatter plot of table 1. The mean value of maximum load below the xiphoid
was larger than that on the back. Fig. 3 showed the results of the mean value in the two loading
modes. The confidence interval of the mean was defined as 90% and we got P<0.1, which indicated
that there was a great difference between the two modes. Thus we could conclude that the bearing
capacity below the xiphoid was stronger than that on the back. The number 2 pigeon stood out the
five pigeons, so we eliminated it and got a new comparison of mean value which is shown in Fig. 3
Fig. 1. Pigeons in bearing capacity test.
Left is carrying a load in front of chest (below the xiphoid) and the right is on the back.
124 Advances in Bionic Engineering
The pigeon’s maximum bearing capacity during takeoff was about 25～27% of its body mass
and with this load it could fly to a 1.4 meter high rod. The measuring way was different from what
Peng Chai described in his paper . The load he used was a 76 cm long thread with added weights.
Here we only used a small bag filled with weights instead because the pigeon may trip over by a
long tread. During the experiment some tips were noted. For example, pigeons must be given freely
Fig. 2. The scatter plot in the two
loading modes. The mean value
below the xiphoid and on the
back was 25.40% and 20.09%
respectively. The blue lines
represent mean value and the red
lines represent the standard
deviation. The solid circles and
rectangles represent the five
pigeons’ ratio data as showed in
Fig. 3. The results of mean value comparison using one-way ANOVA process.
A: the difference between groups with five original data, P<0.1. B: the difference between
groups with number 2 pigeon eliminated, P<0.05. Either A or B could showed that the rational
placement of loading was below the xiphoid of the pigeon body.
Applied Mechanics and Materials Vol. 461 125
flying time during the experiment period for not weakening their flying abilities or flying
willingness with long time captivity. Do not injure the wings and feathers when loading weights to
the pigeon. Pigeon should be allowed to rest after each flight for takeoff to recover energy
consumed during this process.
As to the loading mode, we only chose the back and chest (below the xiphoid), on account that
other parts such as foot and head may not bear heavy load as the two locations tested using common
sense. In the test with loading below xiphoid, bag filled with weights was hung on a little hook (Fig.
1A), so it cannot avoid swaying during takeoff. This may reduce the lifting capacity during takeoff,
but the results showed that even in this case the bearing capacity was much better than that on the
back. Therefore, we can estimate that if the bag clings to the pigeon’s chest as shown in Fig. 1B, the
differences between the bearing capacities of the two modes will be more obvious. The bearing
capacity below the xiphoid was 5.31 percent (mean value) higher than that on the back. This may be
due to the height of the center of gravity with load on the back thereby the stability during takeoff
was reduced significantly. It’s also possible that the wing beat action was slightly blocked due to the
bag’s size on the back. The area that can completely keep clear of the wings on the back was very
small, the size of the bag should be designed adequately.
The greatest hallmark that sets a pigeon-robot apart from a common mechanical robot is its
strong elusiveness. We could refer to the rational loading mode and maximum load of a pigeon
when equipping the bird with some specific devices. In future, more details about take off with load
could be recorded with a high speed camera [11-12]. This was we can learn more about the lift
production, aerodynamics, and hydromechanics during takeoff of pigeons, which may be expected
to be helpful to design of a bionic flying micro-robot.
I would like to thank my workmates Jingdan Shao for her assistance with this experiment and
Lei Shang for his help in building an experimental arena. This research was partially supported by
China Postdoctoral Science Foundation (2012T50497).
 X.C Su, R.T Huai, J.Q Yang, H. Wang, C.Z. Lv, Brain mechanism and methods for
robot-animal motor behavior control, Sci. China Inform. Sci. 42 (2012) 1130-1146.
 E.S. Hawley, E.L. Hargreaves, J.L. Kubie, B. Rivard, R.U. Muller, Telemetry system for
reliable recording of action potentials from freely moving rat, Hippocampus. 12(4) (2002) 505-513.
 A.L. Vyssotski, A.N. Serkov, P.M. Itskov, G. Dell'Omo, et al., Miniature neurologgers for
flying pigeons: multichannel EEG and action and field potentials in combination with GPS
recording. J. Neurophysiol. 95 (2006) 1263-1273.
 T. Ativanichayaphong, J.W. He, C.E. Hagains, Y.B. Peng, et al., A combined wireless neural
stimulating and recording system for study of pain processing, J. Neurosci. Methods. 170(1) (2008)
 P.M. Hughes, J.M. Rayner, Addition of artificial loads to long-eared bats plecotus auritus:
handicapping flight performance, J. Exp. Biol. 161(1) (1991) 285-298.
 P. Chai, J.S. Chen, R. Dudley, Transient hovering performance of hummingbirds under
conditions of maximal loading, J. Exp. Biol. 200(5) (1997) 921–929.
 R.L. Nudds, D.M. Bryant, Consequences of load carrying by birds during short flights are found
to be behavioral and not energetic, Am. J. Physiol. Regul. Integr. Comp. Physiol. 283(1) (2002)
126 Advances in Bionic Engineering
 J.H. Marden, Maximum lift production during takeoff in flying animals, J. Exp. Biol. 130(1)
 P.G. Tickle, M.F. Richardson, J.R. Codd, Load carrying during locomotion in the barnacle
goose (Branta leucopsis): The effect of load placement and size. Comp. Biochem. Physiol. A:
Physiol. 156(3) (2010) 309–317.
 R.H. Gundlach, The visual acuity of homing pigeons. J. Comp. Psychol. 16(3) (1933) 327-341.
 J.R. Usherwood, T.L Hedrick, C.P. McGowan, et al., Dynamic pressure maps for wings and
tails of pigeons in slow, flapping flight, and their energetic implications, J. Exp. Biol. 208(2) (2005)
 A.M. Berg, A.A. Biewener, Wing and body kinematics of takeoff and landing flight in the
pigeon (Columba livia), J. Exp. Biol. 213(10) (2010) 1651-1658.
Applied Mechanics and Materials Vol. 461 127