Mandyam Srinivasan

From visual guidance in insects to autonomous aerial vehicles

Mandyam V. Srinivasan

Queensland Brain Institute, University of Queensland
and
ARC Centre of Excellence in Vision Science

Investigation of the principles of visually guided flight in insects is offering novel, computationally elegant solutions to challenges in machine vision and robot navigation.

Insects, in general, and honeybees, in particular, perform remarkably well at seeing and perceiving the world and navigating effectively in it, despite possessing a brain that weighs less than a milligram and carries fewer than 0.01% as many neurons as ours does. Although most insects lack stereo vision, they use a number of ingenious strategies for perceiving their world in three dimensions and navigating successfully in it.

The talk will describe a series of experiments which reveal that flying insects perceive the world in three dimensions and navigate safely in it by using cues derived from image motion, rather than complex stereo mechanisms. For example, distances to objects are gauged in terms of the apparent speeds of motion of the objects' images, rather than by using complex stereo mechanisms. Objects are distinguished from backgrounds by sensing the apparent relative motion at the boundary. Narrow gaps are negotiated safely by balancing the apparent speeds of the images in the two eyes. The speed of flight is regulated by holding constant the average image velocity as seen by both eyes. This ensures that flight speed is automatically lowered in cluttered environments, and that thrust is appropriately adjusted to compensate for headwinds and tail winds. Visual cues are also used to compensate for crosswinds. Bees landing on a horizontal surface hold constant the image velocity of the surface as they approach it, thus automatically ensuring that flight speed is close to zero at touchdown. Bees approaching a vertical surface hold the rate of expansion of the image of the surface constant during the approach, again ensuring smooth docking. Foraging bees gauge distance flown by integrating optic flow: they possess a visually-driven "odometer" that is robust to variations in wind, body weight, energy expenditure, and the properties of the visual environment.

We have been using some of the insect-based strategies described above to design, implement and test biologically-inspired algorithms for the guidance of autonomous terrestrial and aerial vehicles. Maneuvers such as visually stabilized hover, corridor and gorge navigation, attitude stabilization and terrain following will also be discussed in the talk.

Insects accomplish all of their navigational tasks without relying on detailed terrain maps, or external aids such as GPS. Thus, the principles uncovered by these studies could be of relevance to autonomous navigation in building interiors, in underwater environments, and in space exploration, and may be of interest in designing the next generation of autonomous vehicles, and in formulating future DARPA Grand Challenges.

Mandyam Srinivasan

Dr. Mandyam Srinivasan is with the University of Queensland's Queensland Brain Institute, and the ARC Centre of Excellence in Vision Science. After receiving his Ph.D. at Yale University in Engineering and Applied Science in 1976, he joined the Australian National University's Institute of Advanced Studies where he held a joint appointment as a Research Fellow in the Research Schools of Physical Sciences and Biological Sciences. During this time he worked on the electrophysiological analysis of visual information processing in insects, and on the development of mathematical models of movement detection and optimal image encoding. During 1982-85 he was a Visiting Assistant Professor at the University of Zurich, where he worked on the behavioral analysis of visual processing in honeybees. Srinivasan returned to the Australian National University in 1985, where he was promoted to full professor in 1993.

Dr. Srinivasan's research focuses on understanding the principles of vision and navigation in small animals with relatively simple nervous systems, and on the application of these principles to the design of novel, biologically-inspired algorithms for machine vision, robotics and autonomous vehicles.

Dr. Srinivasan has published over two hundred journal papers and reviews, including several in high-impact journals such as Nature, Science and PNAS. Since 2000 he has delivered numerous keynote/plenary/named lectures at international conferences. He was awarded a D.Sc. in Neuroethology by the Australian National University in 1993, an Honorary Doctorate by the University of Zurich in 2001, the Australian Centenary Medal in 2003, the Australia Prime Minister's Science Prize in 2006, and the U.K. Rank Prize in Optoelectronics in 2008. Srinivasan was elected to the Fellowship of the Australian Academy of Science in 1995, to the Fellowship of the Royal Society of London in 2001, and to the Academy of Science for the Developing World in 2005. He was awarded a five-year Federation Fellowship by the Australian Research Council in 2002, and a five-year Queensland Premier's Fellowship in 2007. Srinivasan has served/ is serving on the Editorial Boards of Vision Research, the Public Library of Science (PLOS), the Journal of Comparative Physiology (A), the Journal of Insect Physiology, the Australian Journal of Intelligent Information Processing Systems, and the IEEE Transactions on Pattern Analysis and Machine Intelligence. Srinivasan's research has received substantial support from the Australian Research Council, the Queensland State Government, the Australian Defense Science and Technology Organization (DSTO), the U.S. Defense Advanced Research Projects Agency (DARPA), the U.S. Office of Naval Research (ONR), the U.S. Air Force (AFOSR), the U.S. National Aeronautics and Space Administration (NASA), and the US Army (ARO).