Haoxiang Cao

Navigation research on outdoor miniature reconnaissance robot

Miniature robot has becoming a preferable choice for outdoor reconnaissance task due to its flexibility. However, miniature robot tends to have poor navigation ability because of the limited perception ability coming with the constrained space. This paper focus on the navigation problem of outdoor miniature reconnaissance robot with limited sensing ability in unknown obstacles environment. In this paper, Global Position System (GPS), magnetometer and infrared distance sensor based artificial potential filed method is adopted to navigation robot go through obstacles. Meanwhile, local minimum trapped detection has been analyzed. Simulations and experiment proves the effectiveness of proposed method.

Actively-compliant locomotion control with the hydraulic quadruped robot on rough terrain

This paper is authored to describe a control framework that is designated for hydraulically actuated quadruped robot to trot on rough terrain. In order to succeed in trotting on rough terrain, two controllers are synthesized: i) Dual Length Linear Inverted Pendulum Method (DLLIPM), ii) Active Compliance Control. The first controller computes the hydraulic quadruped robots trajectory, which not only effectively reduces the energy dissipation, but also promotes the workspace utilization. The second controller, in the meantime, utilizes the force sensors which are located at the bottom of the feet to calculate the joint displacements that are associated with ground reaction force errors, using admittance blocks. In addition to position feedback, these joint displacements are inserted to the position control loop and then updates the orientation input. In doing so, the hydraulic quadruped robot can perform the given locomotion task in an actively-compliant manner. Using the proposed frame work, the overall control performance is tested by hydraulic quadruped robot on rough terrain via simulation and the results turn out to be positive.