Mostafa Ajallooeian

Learn More
On the hardware level, we are proposing and testing a bio-inspired quadruped robot design (Oncilla robot), based on lightweight , compliant, and three-segmented legs. Our choice of placing the compliance such that it is spanning two joints enforces a non-linear spring stiffness. Based on the SLIP-model assumption, we compare progressive and de-gressive(More)
We present the design of a novel compliant quadruped robot: Cheetah-cub, and a series of locomotion experiments with fast trotting gaits. The robot's leg configuration is based on a spring-loaded, pantograph mechanism with multiple segments. A dedicated open loop locomotion controller was derived and implemented. Experiments were run in simulation and in(More)
In general, imitation is imprecisely used to address different levels of social learning from high level knowledge transfer to low level regeneration of motor commands. However, true imitation is based on abstraction and conceptualization. This paper presents a conceptual approach for imitation learning using feedback cues and interactive training to(More)
—We present a modular controller for quadruped locomotion over unperceived rough terrain. Our approach is based on a computational Central Pattern Generator (CPG) model implemented as coupled nonlinear oscillators. Stumbling correction reflex is implemented as a sensory feedback mechanism affecting the CPG. We augment the outputs of the CPG with virtual(More)
— We present a general approach to design modular controllers for limit cycle locomotion over unperceived rough terrain. The control strategy uses a Central Pattern Generator (CPG) model implemented as coupled nonlinear oscillators as basis. Stumbling correction and leg extension reflexes are implemented as feedbacks for fast corrections, and model-based(More)
We present a general family of nonlinear phase oscillators which can exhibit arbitrary limit cycle shapes and infinitely large basins of attraction. This general family is the superset of familiar control methods like PD-control over a periodic reference, and rhythmic Dynamical Movement Primitives. The general methodology is based on morphing the limit(More)
In this paper, we propose a fast algorithm for gesture recognition based on the saliency maps of visual attention. A tuned saliency-based model of visual attention is used to find potential hand regions in video frames. To obtain the overall movement of the hand, saliency maps of the differences of consecutive video frames are overlaid. An improved(More)
1 Motivation We are motivated to build simple controllers for quadruped robots to locomote over unperceived moderately difficult rough terrain at moderately fast speeds. The presented approach here does not need force sensing at feet, and does not need information about the mass properties of the robot like inertia tensors, so it is apt for relatively cheap(More)