Robot Foot Contact Detection Using Joint Torque Dynamics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting foot/ground contact in legged robots are impractical, costly, or dependent on inaccurate model parameters, and fail to account for leg mass and speed, leading to instability and control issues.
Innovation Solution
A method using torque sensors or motor current to estimate joint torques, combined with force Jacobian matrices, to calculate the rate of change in foot forces, determining contact through a caching queue and threshold comparison.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If additional sensors (e.g., piezoelectric strain gauge) are installed on the sole of each foot, then foot/ground contact detection accuracy is improved, but manufacturing cost and device complexity increase
Solution Approach 1:
The patent extracts the contact detection function from dedicated sole-mounted sensors and relocates it to existing torque sensors or motor current measurements at the joint level. By measuring joint torques and using dynamic models to infer foot/ground contact status, the system eliminates the need for additional sensors on the foot soles, thereby reducing device complexity while maintaining detection accuracy
Solution Approach 2:
The patent makes existing torque sensors and motor current measurements serve dual purposes: their primary function for joint control and an additional function for foot/ground contact detection. This multi-functionality approach allows the system to perform contact detection without adding dedicated sensors, thus reducing overall device complexity while maintaining measurement precision
2Device complexity
If torque sensors or motor current are used to monitor foot force, then device complexity is reduced, but measurement precision and response speed deteriorate due to model parameter accuracy and lag
Solution Approach 1:
The patent transitions from static or quasi-static models to a dynamic model that explicitly accounts for leg mass, acceleration, and velocity. The dynamic model uses real-time joint torque measurements combined with leg kinematic parameters (mass, length, acceleration) to calculate foot forces, thereby improving measurement precision and reducing lag compared to static observer methods
Solution Approach 2:
The patent changes the modeling approach by incorporating leg mass parameters and dynamic acceleration terms into the force calculation model. By using dynamic equations that include mass-dependent terms (M(q)q̈ + C(q,q̇)q̇ + g(q)), the system adapts to different leg configurations and speeds, improving measurement precision across varying operating conditions
3Device complexity
If static Jacobian matrix is used to estimate plantar force, then device complexity is reduced, but applicability deteriorates when leg mass cannot be ignored
Solution Approach 1:
The patent replaces the static Jacobian matrix approach with a dynamic model that incorporates leg mass, acceleration, and velocity parameters. The dynamic model (M(q)q̈ + C(q,q̇)q̇ + g(q) = J(q)ᵀF) remains computationally tractable while being applicable to robots with varying leg masses and configurations, thereby improving versatility without significantly increasing calculation complexity
Data Source
AI summary
A method for detecting contact of a swinging leg of a robot with ground includes: obtaining a torque on each joint of the swinging leg when the robot is in a swing phase; estimating a force on a foot of the swinging leg by using a force Jacobian matrix based on the torque on each joint of the swinging leg, and calculating a rate of change of force of the foot in a vertical direction according to the force on the foot; and determining that the swinging leg has contacted the ground in response to a preset consecutive number of values of the rate of change of force being greater than a preset threshold.