Ambarish Goswami, Ph.D.


Ambarish Goswami, Ph.D.

Principal Scientist
Honda Research Institute, US
Mountain View, CA, USA



Brief Biography

Ambarish Goswami has been with Honda Research Institute in California, USA, for the past eleven years, where he is currently a Principal Scientist. His field is dynamics and control, and his main research is in the application areas of humanoid robots, assistive exoskeletons and vehicles. He received the Bachelor's degree from Jadavpur University, Kolkata, India, the Master's degree from Drexel University, Philadelphia, PA, and the Ph.D. degree from Northwestern University, Evanston, IL, all in Mechanical Engineering.

Ambarish Goswami's Ph.D. work, under Prof. Michael Peshkin, was in the area of automated assembly and robot-assisted surgery. For four years following his graduation he worked at the INRIA Laboratory in Grenoble, France, as a member of the permanent scientific staff (Charge de Recherche). He became interested in human walking and in biomechanics while working on "BIP," the first anthropomorphic biped robot in France. This interest in gait study subsequently brought him to Prof. Norm Badler's Center for Human Modeling and Simulation at the University of Pennsylvania, Philadelphia, as an IRCS Fellow, and a three year position as a core animation software developer for 3D Studio Max at Autodesk. Ambarish has held visiting researcher positions at the Ohio State University and the University of Illinois at Urbana-Champaign for short periods. 

Ambarish is an originator of the widely used Compass Gait Model of biped walking which he has used to elucidate gait stability, period-doubling and chaos. Ambarish has also contributed to the understanding of humanoid robot balance and has introduced the Foot Rotation Indicator (FRI) point. He has addressed topics such as push recovery and fall control. Ambarish has more than 70 publications with a total of more than 3700 Google Scholar citations; he has twelve patents. Ambarish is in the Editorial Board of International Journal of Humanoid Robotics (World Scientific) and Robotica (Cambridge University Press), and is one of the Editors-in-Chief of the forthcoming Springer Handbook of Humanoid Robotics. 

Ambarish is an ASME Fellow (2013).


Fall Control of Humanoid Robots

Human-centered robots, first of all, must be completely safe. This motivation has recently attracted a great deal of attention from the research community and resulted in the development of innovative technologies including lightweight and compliant manipulator arms. Autonomous walking robots have a unique and serious safety issue resulting from a loss of balance and fall. Self-damage and unintentional human injury from a fall are major reasons that humanoid robots are not allowed to move freely in human environments.

Balance is one of the oldest robotics topics that continue to appeal to researchers of the present day. However, keen interest in humanoid robot balance sometimes makes us overlook the consequences of a balance failure. Although fall appears to be a rare event in the life of a humanoid robot, its occurrence is virtually unavoidable and its consequences can be disastrous.

A falling robot is an underactuated system that rapidly gains speed under gravity. The time to act is very short. How can one intervene? In this respect, we will describe our work on humanoid robot fall strategy which tries to modify the robot's fall direction in order to avoid hitting a person or an object in the vicinity. Our approach is based on intelligent foot placement as well as a method called "inertia shaping" which is aimed at controlling the centroidal inertia of the robot. When the chances of hitting another person is low, the robot attempts to fall in a manner so as to reduce material damage to itself. We demonstrate our results through the simulation and hardware experiments.