Alborz Aghamaleki Sarvestani received his BSc and MSc degrees in Mechanical Engineering from Shiraz University. His MSc thesis was " Design, manufacturing, dynamic modeling and control of a lower extremity exoskeleton with validation by experimental data".
He joined Dynamic Locomotion Group in 2017. His main focus is on designing biped base on the blueprint from nature and especially birds to investigate basic research question in bird locomotion. In order to fulfill this objective, he works on the various leg mechanisms, actuators, and instrumentation.
He is generally interested in the leg design, mechanisms design and analyzes, bio-inspired robots, multibody dynamics, robotics, mechatronics.
Animals outperform robotic walking machines even though they have restrictions in actuator power density, and limited control loop speed. Despite animals only having access to sparse feedback information, due to limited nerve speed, they run faster, more robust and efficient compared to robots. We expect an 'intelligence' in...
In reinforcement learning, tasks that are difficult to learn are often made more amenable by shaping the reward (cost) landscape. This is typically done by adjusting the reward signal $R$ in the Markov Decision Process, composed of $(S,A,P,R,\gamma)$, where $S$ is the state-space, $A$ is the action-space, $P$ is the probabiolity tra...
We investigate the functional demands of bipedal running with a focus on the concept of postural stability.
Bipedalism has evolved in many lineages ranging from reptiles, avians, theropods to primates. These lineages exhibit diversified morphologies (e.g. limb segment length, mass, orientation), which yield to different lo...
In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA) 2018, pages: 5076-5081, IEEE, International Conference on Robotics and Automation, May 2018 (inproceedings)
Learning instead of designing robot controllers can greatly reduce engineering effort required, while also emphasizing robustness. Despite considerable progress in simulation, applying learning directly in hardware is still challenging, in part due to the necessity to explore potentially unstable parameters. We explore the of concept shaping the reward landscape with training wheels; temporary modifications of the physical hardware that facilitate learning. We demonstrate the concept with a robot leg mounted on a boom learning to hop fast. This proof of concept embodies typical challenges such as instability and contact, while being simple enough to empirically map out and visualize the reward landscape. Based on our results we propose three criteria for designing effective training wheels for learning in robotics.
Iranian Journal of Science and Technology, Transactions of Mechanical Engineering, 40, pages: 77–85, Springer International Publishing, March 2016 (article)
Gas and liquid pipelines surround us. To ensure reliable product delivery and to maintain pipeline integrity, asset managers should consider routine pipeline inspection and holistic management programs to extend pipeline life and prevent risk. Therefore, pipe inspection robots are of special interest to industries. In this paper, we present a new and simple locomotion strategy for an out-pipe inspection robot which can provide adjustable tractive force and can also be utilized to support active diameter adaptability. The advantages proposed by this design include simplicity, low manufacturing costs, online inspection capability and short operational time. Here a dynamic model of the robot is presented with the required assumptions. The mathematical model of 2-DOF robot is obtained using the well-known Lagrange equation. Modeling and simulations were conducted to test the validity and practicality of the proposed design and strategies. The prototype has successfully traveled along a pipe of 20 cm diameter. The results obtained from our dynamic model are then validated by experimental data.
Our goal is to understand the principles of Perception, Action and Learning in autonomous systems that successfully interact with complex environments and to use this understanding to design future systems