Shape-Based Compliant Control of the Robot

Shape-Based Compliant Control of the Robot

Jun 7, 2016 @ 20:59 | CMU | USA | 

Our real world is full of unstructured and complex environments and natural creatures move through these environments very smoothly with adaptability and robustness. Scientists are trying to understand deeply that how natural creatures are controlling their motions on such complex terrains and want to make robots that can mimic natural motions? Nature has lots of locomotion techniques (Bipedal, Quadruped, serpentine etc) and control strategies based on locomotion to navigate in complex environments.

Researchers are also trying different locomotion techniques to run the robots on rough terrains. Biorobotics lab, CMU designed shape-based compliance in locomotion and resultant shape-based controller produces behaviors that enable robots to robustly feel their way through unknown environments. The Researchers are presenting their work and finding in June at RSS 2016.

One way to navigate in complex environment is hyper redundant robot with high degree of freedom. Having high degree of freedom is both a blessing and a curse. A mechanism with a large number of degree of freedom can better comply to and therefore better move in complex environments. Yet, possessing high degree of freedom is only an advantage if the system is capable of coordinating to achieve desired goals in real-time.

Related articles:

But higher level path planning and motion controlling becomes difficult as degree of freedom increases. Their work supports the belief that a middle layer of abstraction between conventional planning and control is needed to enable robust locomotion of articulated systems in complex terrains. The basis for this abstraction is the notion that a system’s shape can be used to capture joint to joint coupling and provide an intuitive set of controllable parameters that adapt the system to the environment in real time.

Researchers present a generalizable framework that specifies desired shapes in terms of shape functions. They show how shape functions can be used to link low-level controllers to high-level planners in a compliant control framework that directly controls shape parameters and tested this framework to control two separate locomotion, a snake-like and a hexapod like.

Decoupling planning and control for locomotion in complex terrains does not, in general, produce robust real-time solutions. A planner can generate a motion plan, and then a closed-loop controller can execute it, but any slight uncertainly in modeling can easily cause the system to fail. In other words, decoupled planning/control solutions tend to be brittle. To address this, Researchers fuse planning and control with a coherent middle layer that adds robustness to and decreases the complexity of coordinating many degrees of freedom during locomotion. Shape functions are presented as the element that helps define this middle layer for articulated systems. A shape function

1) Analytically encodes the interdependent joint-to-joint motions necessary to produce desired behaviors in nominal locomotion,

2) Provides sets of parameters that can be controlled to adapt mechanisms to complex environments while moving through them, and

3) Offers low-dimensional, straight-forward connections to higher-level planners.

In terms of adapting locomotion in unstructured terrains, this work shows that admittance control can be specified in terms of shape parameters, which are the parameters used to define shape functions. By mapping joint torques into equivalent forces on shape, shape-based compliant controllers autonomously modify the system’s shape to better comply to the environment during locomotion.

The shape is an important component in creating a middle layer that links high-level motion planning to low-level control in the robust locomotion of articulated systems in complex terrains.

 Reference: “Shape-Based Compliance in Locomotion”, Matt Travers, Julian Whitman, Perrin Schiebel, Dan Goldman, and Howie Choset. paper

Leave a Reply

Your email address will not be published. Required fields are marked *