Skip to main content

Center of Mass and Dynamic Balance

Learning Objectives​

  • Define the concepts of Center of Mass (CoM) and Zero Moment Point (ZMP).
  • Understand how CoM and ZMP are used to achieve static and dynamic balance in bipedal robots.
  • Recognize strategies for controlling a robot's balance during locomotion.

Core Concepts​

Maintaining balance is the single most critical challenge for bipedal robots. Humans unconsciously perform complex calculations to keep their Center of Mass (CoM) aligned and their Zero Moment Point (ZMP) within their support area. Robotics aims to explicitly model and control these same physical principles.

Center of Mass (CoM)​

The Center of Mass (CoM) is the unique point where the weighted relative position of the distributed mass of an object sums to zero. It's the average position of all the mass in a system. For a bipedal robot, controlling the CoM is fundamental to balance.

  • Static Balance: A robot is in static balance if its CoM projection lies within its support polygon (the convex hull formed by the contact points of its feet on the ground). This is easy to achieve when standing still.
  • Dynamic Balance: During walking or dynamic movements, the CoM projection often moves outside the support polygon. The robot must continuously shift its CoM and adjust its foot placement to maintain balance over time.

Zero Moment Point (ZMP)​

The Zero Moment Point (ZMP) is a concept crucial for understanding and controlling the balance of multi-limbed robots. It is defined as the point on the ground where the total moment (torque) of all forces acting on the robot (gravity, inertia, and ground reaction forces) is zero.

  • Stability Criterion: For a robot to remain upright (without falling over), its ZMP must lie within its support polygon.
    • If ZMP is inside the support polygon: The robot is stable.
    • If ZMP is outside the support polygon: The robot will fall.
  • Control Strategy: Robot control systems often plan trajectories that ensure the ZMP remains within the support polygon, or rapidly bring it back inside if disturbances occur. This involves precise control of joint torques and foot placements.

Strategies for Balance Control​

  1. Ankle Strategy: Small adjustments made by rotating the body about the ankles, similar to how humans sway slightly.
  2. Hip Strategy: Larger adjustments involving bending at the hips and knees, allowing for a wider range of motion to recover balance.
  3. Step Strategy: If other strategies fail, the robot takes a step to create a new support polygon under its falling CoM.
  4. Momentum Control: Utilizing the robot's own inertia and momentum to aid in balance and efficient movement, especially during dynamic gaits.

Hands-On Exercise​

Exercise: Specifying ZMP Control for a Simple Balancing Act

  1. Specification (SDD Phase 1): You are tasked with keeping a simulated bipedal robot balanced on a single foot for 5 seconds.

    • Task: Define the support polygon in this scenario.
    • Task: How would you measure the ZMP of the robot in simulation (conceptually)?
    • Task: What sensor feedback would be essential for controlling the ZMP (e.g., IMU for orientation, force sensors in the foot)?
    • Task: Propose a high-level control strategy to keep the ZMP within the single-foot support polygon. Consider how the robot's body parts (e.g., arms, other leg) could be used to shift the CoM.
    • Task: Define the success criterion: "The robot remains balanced on one foot for 5 seconds without its ZMP leaving the support polygon."

  2. Disturbance Handling (SDD Phase 2): If a small external force momentarily pushes the robot, how would your control strategy react to bring the ZMP back into the support polygon? Describe the steps.

Summary​

The concepts of Center of Mass and Zero Moment Point are fundamental to achieving stable locomotion and balance in humanoid robots. By carefully planning and controlling the robot's dynamics to keep its ZMP within the support polygon, engineers can design systems that walk, run, and interact with the world with increasing agility, bridging the gap between human and robotic movement.