By solving as a nonlinear program (NLP) with complementarity constraints, a multi-limbed robot is capable of transiting between the ground and wall with different contact sequences given different environment setups.

A new type of quadruped robot with reduced degrees of freedom (DoF) requires minimal-torque actuators to achieve high-payload locomotion. Design is optimized via nonlinear programming (NLP).

Motion planning for multi-limbed robots to climb up between walls is decoupled into two parts with distinct physical meaning, the first part is formulated as MICP to solve contact positions and torso postures while the second part solves a series of convex optimization to get feasibile contact forces.

Designed a novel low temperature apparatus including three parts, detector carrying device, device to build low temperature environment and device to build vacuum.