Jaehyun Shim,Carlos Mastalli,Thomas Corbères,Steve Tonneau,Vladimir Ivan,Sethu Vijayakumar
Abstract:State-of-the-art approaches to footstep planning assume reduced-order dynamics when solving the combinatorial problem of selecting contact surfaces in real time. However, in exchange for computational efficiency, these approaches ignore joint torque limits and limb dynamics. In this work, we address these limitations by presenting a topology-based approach that enables model predictive control (MPC) to simultaneously plan full-body motions, torque commands, footstep placements, and contact surfaces in real time. To determine if a robot's foot is inside a contact surface, we borrow the winding number concept from topology. We then use this winding number and potential field to create a contact-surface penalty function. By using this penalty function, MPC can select a contact surface from all candidate surfaces in the vicinity and determine footstep placements within it. We demonstrate the benefits of our approach by showing the impact of considering full-body dynamics, which includes joint torque limits and limb dynamics, on the selection of footstep placements and contact surfaces. Furthermore, we validate the feasibility of deploying our topology-based approach in an MPC scheme and explore its potential capabilities through a series of experimental and simulation trials.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is: When existing footstep - planning methods select contact surfaces in real - time, they usually assume a simplified dynamic model to solve the combinatorial problem. However, in order to improve computational efficiency, these methods ignore joint - torque limits and limb dynamics. This results in that when walking on complex terrains, the robot may not be able to accurately track the predetermined footstep positions and cannot fully utilize its range of motion.
To solve these problems, this paper proposes a topology - based method that enables model predictive control (MPC) to plan whole - body motion, torque commands, footstep positions and contact surfaces simultaneously. Specifically, the article introduces the following improvements:
1. **Consider a comprehensive dynamic model**: Different from existing methods, this method not only considers the simplified model of the robot, but also takes into account joint - torque limits and limb dynamics, thus more accurately simulating the real behavior of the robot.
2. **Real - time selection of contact surfaces and footstep positions**: By introducing the concept of winding number from topology, this method can automatically select the best contact surfaces and footstep positions in a real - time environment.
3. **Optimize the combinatorial problem**: By creating a continuous cost function to evaluate candidate contact surfaces, this method can select the optimal solution among all candidate contact surfaces without the need to pre - define the footstep sequence.
The advantage of this method is that it can ensure the stability and flexibility of the robot on complex terrains while maintaining computational efficiency. In addition, the author verifies the effectiveness of this method through a series of experiments and simulations, demonstrating its application potential on actual robots.
### Formula Summary
- **Electric potential formula**:
\[
V_E(r)=\frac{Q}{4\pi\varepsilon_0|r - p|}
\]
where \(Q\) is the charge of the particle, \(\varepsilon_0\) is the vacuum permittivity, \(r\) is the evaluation point, and \(p\) is the position of the charged particle.
- **Line - segment electric potential integral**:
\[
p_i=\int_0^1\frac{1}{|a_i+(b_i - a_i)s-\hat{p}|}ds
\]
where \(a_i, b_i\in\mathbb{R}^2\) are the endpoints of the line segment, and \(\hat{p}\in\mathbb{R}^2\) is the position of the robot's foot on the horizontal plane.
- **Winding number formula**:
\[
\text{wind}(\gamma,\hat{p})=\frac{1}{2\pi}\sum_i\text{atan2}(c_i,d_i)
\]
where \(c_i = [(a_i-\hat{p})\times(b_i-\hat{p})]_z\) and \(d_i=(a_i-\hat{p})\cdot(b_i-\hat{p})\).
- **Contact - surface penalty function**:
\[
\ell_{pCk}=
\begin{cases}
0 & \text{if }\sum_j V_{\gamma j}=0\\
\sqrt{\sum_j N_j\sum_j V_{\gamma j}} & \text{if }\sum_j\text{wind}(\gamma_j,\hat{p}_{Ck})<\frac{1}{2}\\
0 & \text{if }\sum_j\text{wind}(\gamma_j,\hat{p}_{Ck})\geq\frac{1}{2}
\end{cases}
\]
Through these formulas and methods, this paper successfully overcomes the limitations in existing footstep - planning methods and provides a new solution for the autonomous navigation of robots on complex terrains.