Publicity information for:

K. Peterson, P. Birkmeyer, R. Dudley, R. S. Fearing 

A wing-assisted running robot and implications for avian flight evolution, Bioinspiration and Biomimetics, October 18, 2011 6 046008

Freely available online from Tuesday, October 18th at:

UC Berkeley researchers have recently developed DASH+Wings, a small hexapedal winged robot that uses flapping wings to increase its locomotion capabilities. Based on the Dynamic Autonomous Sprawled Hexapod (DASH), the robot was developed to study the effects of flapping wings on terrestrial locomotion. The impact of the flapping wings on locomotion was quantified by measuring the speed across horizontal surfaces, the maximum incline the robot could ascend, and the glide performance in free flight. We also examined three control configurations provided by wing removal, the use of inertially similar lateral spars, and passive rather than actively flapping wings. Our results showed that flapping wings provide an advantage over the control experiments in nearly all facets of locomotion, improving the horizontal speed, attainable incline, and glide slope.

Noticing that most prior theories on avian flight evolution are based on incomplete fossil records and theoretical modeling, we determined to find out if a hybrid robot could shed some light on the matter. Prior theoretical models predicted that a ground-dwelling animal would need to triple its running speed to allow for takeoff. While the winged robot did improve its terrestrial capabilities, it fell short of the necessary speed-up required to enable flight. Combined with new fossil evidence, we concluded that our robot lends more indirect evidence to the theory that flight evolved from tree-dwelling gliders. Perhaps more importantly, we believe that these experiments demonstrate the feasibility of using robot models to test hypotheses of flight origins, and hope to continue to use robotic models to illuminate the question of avian flight evolution.

Point of Contact
Prof. Ronald S. Fearing, UC Berkeley; 510-642-9193;
ronf @ eecs . berkeley . edu
Kevin Peterson, Ph.D. Candidate, UC Berkeley; 510-859-4457; kevincp @ eecs . berkeley . edu

This work was supported by the NSF Center of Integrated Nanomechanical Systems (PB) and the United States Army Research Laboratory under the Micro Autonomous Science and Technology Collaborative Technology Alliance (KP and RF) and NSF IOS-0837866 (RD).

A dynamic hexapedal robot with flapping wings (DASH+Wings) for (A) flapping wings configuration; (B) legs-only configuration and (C) inertial spars configuration. The fourth configuration (wings passive) is as in (A) but with the wing drive disconnected.
DASH+Wings jpg (1.4 MB) DASH jpg (1 MB) DASH+Inertial Spars jpg (1.1 MB)
(A) (B) (C)

Full resolution videos can be downloaded here:
Horizontal Running
Maximum Incline Attainable
Wing Added Stability
Robot motion (10X slowed)

Engineering Design for Improved Performance

Adding flapping wings to a running robot provides several advantages for traversing complex environments. The wings increase the overall thrust of the robot, enabling it to accelerate faster, and ascend steeper inclines. Wing flapping also increases the stability of the robot when running along the ground, reducing the chances that an obstacle will cause the robot to flip over and potentially become incapacitated. Finally, the wings also add aerial stability to the robot, allowing it to glide through the air instead of falling vertically. They also ensure the robot will land on its feet, enabling it to continue its mission.

Implications for the Origins of Bird Flight

Various ideas as to how birds initially evolved flight have been difficult to evaluate given the absence of a detailed transitional fossil record. One possibility is that wings evolved in tree-dwelling, gliding forms that flapped their wings for weight support and maneuverability. For example, precursors to early birds were characterized by flight feathers on all four limbs, and were furthermore characterized by long feathered tails, suggesting that they were gliding animals. Alternatively, early birds may have been runners, moving either horizontally or up inclined structures, such that wing flapping indirectly led to enhanced running performance and ultimately to flight. Adding wings to the DASH robot increased aerodynamic performance in gliding, and also increased running performance, but not up to the speeds necessary to attain takeoff either horizontally or on an incline. These experimental results are thus more consistent with the aerial gliding hypothesis for the origins of bird flight.


Winged robots have a multitude of both civilian and military applications, specifically in reconnaissance, exploration of hazardous areas, and search and rescue. In a situation such as a collapsed building, the area may be too dangerous for humans to move about safely. In this case the robot could enter the building through gaps humans may not be able to fit through and search for survivors. On the current robot, the wings allow for stable descent. It can jump from high areas and stably land on its legs at the bottom and continue searching. The wings also increase running stability as well as the range of inclines the robot can traverse, further increasing the likelihood of a successful mission. Future hybrid robots under development that are capable of taking off and flying will be able to reach more space in an area - when blocked from exploring further on the ground they can take off and search for alternative entry points.


Kevin Peterson, Paul Birkmeyer, and Ronald S. Fearing

Biomimetic Millisystems Laboratory
Department of Electrical Engineering & Computer Science
University of California at Berkeley

Robert Dudley

Animal Flight Laboratory, Department of Integrative Biology
University of California at Berkeley
Smithsonian Tropical Research Institute, Balboa, Republic of Panama