Research

The purpose of this project is to expand the autonomous navigation capabilities for the MIT Cheetah 2 robot. Recent work has led to the development of a high-speed running controller which is capable of running between 0-6.4 m/s without changing any controller parameters. This controller has enabled the development of a higher-level controller to coordinate autonomous jumps to clear obstacles. The cheetah robot is able to jump over obstacles up to 40 cm in height while running at 2.4 m/s. 

Using new design principles and methodologies, we have developed a multi-axis, large force detecting foot sensor for legged robots. This footpad sensor is intended for use on the MIT Cheetah to provide a complete picture of the ground interaction forces that is a necessity in enabling high-speed and dynamic ground locomotion.

In designing a robot, the actuator’s allowable mass and required output torque are determined by the application. However, these requirements still leave a broad design space within which to select motor size and gear ratio. We have developed an actuator design method that enables force control in applications with highly dynamic environmental interactions. The method optimizes the motor selection and gear ratio for high fidelity proprioceptive force control within given actuator weight constraints. We implemented the method in the primary actuators of the MIT Cheetah.

Swing leg retraction (SLR) is a behavior exhibited by humans and animals in which the airborne front leg rotates rearward prior to touchdown.  The Biomimetic Robotics Lab investigates the effects of swing leg retraction on several metrics of running performance to develop intuition for robot controller design.

The MIT Super Mini Cheetah is an inexpensive and lightweight quadrupedal robot that is capable of behaviors such as running, walking, jumping and turning. The design of the Super Mini Cheetah follows many ideals of MIT Cheetah while emphasizing the use of commercial-off-the-shelf components and low-cost rapid manufacturing methods.

The MIT HERMES humanoid robot system is designed for studying whole-body human-in-the-loop control with balance feedback. Inspired by the innate physical control capabilities of humans as well as the capacity for creative learning, we explore the use of the full-body of the human operator as the controller for a humanoid robot.

 

The Biomimetic Robotics Lab investigates the way of using a tail to improve maneuverability of the MIT Cheetah. The research was initiated with the inspiration from videos, showing that the cheetah’s turn is accompanied by a movement of its tail, and some researches in biology, describing that cats or dogs are moving their tail during locomotion. We end up with hypothesizing that a tail may enhance balance of the legged robot. This hypothesis is investigated with three examples.

Taking inspiration from the complementary arrangement of bones and tendons in biology, we use separate elements for compressive and tensile loads to create lightweight structures capable of supporting large moment loads.

The goal of this project is to build a robotic quadruped similar to the MIT Cheetah, but on a much smaller scale and at significantly lower cost.  The robot’s actuators will be  inexpensive brushed DC motors with 18.75:1 gearboxes.  The motors are driven with hobby-grade H-bridge ICs, and controlled by an mbed microcontroller.

Publications