Bipedal Robot Gait Control via Nonlinear Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for controlling bipedal robots lack effective strategies for generating proper foot placements that account for the dynamics of walking motion and actuator limits, leading to inefficient walking and balancing on flat and non-flat surfaces.
Innovation Solution
A control method that includes a foot placement algorithm to compute the Gait Reference Point (GRP) using nonlinear optimization, considering the full dynamics of the robot and ground collisions, and a torque calculator to generate desired joint torques for achieving balanced sagittal-plane walking, which can be extended to three-dimensional walking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional control methods are used for bipedal robots, then the control system is simpler, but the walking efficiency and stability deteriorate
Solution Approach 1:
The control system is segmented into three distinct modules: foot placement algorithm (determining optimal foot positions), gait generator (producing joint trajectories), and torque calculator (computing required torques). This segmentation allows each module to specialize in a specific function, improving overall walking efficiency while maintaining manageable complexity through modular design.
Solution Approach 2:
The foot placement algorithm performs preliminary action by determining optimal foot positions before the gait generator creates joint trajectories. This advance planning of foot placements allows the system to optimize walking efficiency proactively rather than reactively, improving overall system performance.
2Stability of the object's composition
If foot placement is not optimized, then the control system is simpler, but the robot's ability to maintain balance deteriorates
Solution Approach 1:
The foot placement algorithm uses feedback from the robot's current state (joint positions, velocities, and dynamics) to determine optimal foot placements. This feedback mechanism ensures the robot maintains balance by continuously adjusting foot positions based on real-time system state, improving stability through dynamic adaptation.
Solution Approach 2:
The foot placement control incorporates full robot dynamics including ground collisions, making the system dynamically adaptive rather than statically predetermined. This dynamic approach allows the robot to maintain balance on various surfaces by adjusting foot placements based on real-time dynamic conditions.
3Reliability
If actuator limits are not considered in control, then the control algorithm is simpler, but the robot cannot achieve safe and compliant joint control
Solution Approach 1:
The torque calculator performs preliminary anti-action by pre-calculating required torques while explicitly considering actuator limits. This prevents unsafe or non-compliant joint control by ensuring torque commands never exceed actuator capabilities, maintaining reliability through proactive constraint enforcement.
Solution Approach 2:
The control algorithm incorporates actuator limits as explicit parameters in the torque calculation process. By changing the control approach to include these parameter constraints, the system achieves safe and compliant joint control while maintaining algorithmic tractability through proper parameter formulation.
4Measurement precision
If full dynamics including ground collisions are considered, then the foot placement accuracy is improved, but the computational complexity increases
Solution Approach 1:
The dynamics model is segmented into manageable components: robot kinematics, robot dynamics, and ground collision models. This segmentation allows the foot placement algorithm to consider full dynamics for precision while maintaining computational tractability through modular model structure.
Solution Approach 2:
The foot placement algorithm performs preliminary forward simulation to predict the consequences of candidate foot placements before selecting the optimal placement. This preliminary action allows the system to achieve high precision by evaluating multiple options against the full dynamics model, then selecting the best placement in advance.
Data Source
AI summary
A control method that executes a numerical non-linear optimization procedure with forward simulation of the full dynamics of the bipedal robot (including ground collisions) to compute a foothold, which produces the desired center of mass velocity during the next step of the walking cycle. The controller includes a gait generator that outputs desired joint trajectories designed to track a desired foothold provided by the foot placement algorithm. A torque calculator is included to output desired joint torques designed to track the desired joint trajectories provided by the gait generator. Additionally, the controller includes an actuator controller that produces the desired joint torques determined by the torque calculator at each joint on the physical robot, which may be a legged robot with torque-controlled joints. The foot placement algorithm may use nonlinear optimization that is solvable using a mathematical model of the dynamics of the controlled robot.