Hybrid Robotics

Robust grasping, dexterous manipulation, and locomotion skills are the main embodiments of a robot to perform complex mobile manipulation tasks, which are paramount in most of the industrial and service applications. In typical mobile manipulators, these embodiments are realized with dedicated subsystems such as

1. Grasper to handle an object,
2. Manipulator for changing the pose of an object being grasped,
3. Mobile platform for locomotion.

Therefore, the actuators powering the mobile platform cannot be used for manipulation, and the actuators powering the arm joints cannot be used for locomotion. In such robot designs, the mobile platform helps the manipulator to navigate an environment and stays still while the arm performs a manipulation task. The actuators can be better utilized if they can be used interchangeably for locomotion and manipulation tasks, which results in increased versatility and adapting to the task in a novel way.
A few robotic designs found in prior art, have been developed with versatility as the primary design goal where multiple locomotion modes have been achieved at the expense of increased complexity. One of the main focus of this research is to develop a multimodal robotic system that can overcome the need for domain-specific robots. There is a need for such a simple, yet versatile robotic platform in various fields like flexible manufacturing, exploration, search-and-rescue, pipe climbing, warehouse, etc. The novelty of the proposed robot can be broken down into two categories as

1. Multipurpose grasper having the capabilities of shape conformation, within-hand manipulation, and a built-in compact mechanism to vary and estimate the forces at the contact surface,
2. An articulated floating base serial chain having flexible joints equipped with a pair of such graspers to achieve various grasping, manipulation, and locomotion modes.

Optimal control framework has been utilized to find a feasible trajectory to achieve successful locomotion. The mathematical model for the underactuated grasper and the robot has been developed, and the simulation results have been validated with a fabricated prototype.

Journal publications

• Nagamanikandan Govindan and Asokan Thondiyath, "Design and analysis of a multimodal grasper having shape conformity and within-hand manipulation with adjustable contact forces," ASME Journal of Mechanisms and Robotics, vol. 11, no. 5, p. 051012, October 2019, doi: 10.1115/1.4044163.

Patents

• Nagamanikandan Govindan, Sai Sourya Varenya Kovvali, Karthik Chandrasekaran, and Asokan Thondiyath, "A versatile hybrid robotic platform with a pair of multifunctional graspers for multimodal locomotion and grasping," Indian Patent Application No. 201841008257, March 06, 2018.

Conference publications

• Nagamanikandan Govindan, Sai Sourya Varenya Kovvali, Karthik Chandrasekaran, and Asokan Thondiyath, "GraspMan - A Novel Robotic Platform with Grasping, Manipulation, and Multimodal Locomotion Capability," in IEEE International Conference on Robotics and Automation (ICRA), Brisbane, QLD, Australia, 2018, pp. 7354-7359, doi: 10.1109/ICRA.2018.8462970.
• Nagamanikandan Govindan and Asokan Thondiyath, "Improved Planning and Filtering Algorithm for Task-priority Redundancy Resolution in Mobile Manipulation," in Proceedings of the 14th International Conference on Informatics in Control, Automation and Robotics (ICINCO '17), Madrid, Spain, vol. 2, pp. 247--253, 2017, doi: 10.5220/0006402002470253.

Faculty in-charge: Prof. Asokan Thondiyath

Students: Mr. Nagamanikandan Govindan