A. James Clark School of Engineering
Permanent URI for this communityhttp://hdl.handle.net/1903/1654
The collections in this community comprise faculty research works, as well as graduate theses and dissertations.
Browse
3 results
Search Results
Item The Curved Openspace Algorithm and Neuromorphic Mechanisms for Sonar-Based Obstacle Avoidance(2021) Wen, Chenxi; Horiuchi, Timothy TKH; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Bats are known for their ability to pursue a goal while avoiding obstacles in a cluttered environment using ultrasonic echolocation. This dissertation explores two neuromorphic mechanisms involved in such a task: a conductance-based neuron circuit for azimuthal echolocation whose dynamic range can be expanded by power-law compression, and a sonar-based obstacle avoidance algorithm with its spike-latency model implementation. Bats and other mammals use interaural level differences (ILD) to estimate the direction of high-frequency sounds. To compute the ILD of a sound, independent of overall loudness, excitatory and inhibitory synaptic conductances (encoding the left and right amplitudes) are hypothesized to compete in the neurons of the lateral superior olive. Interestingly, this neural model can also accept power-law compressed amplitudes that can allow a much larger range of input signal levels, a common limitation in neural coding. This dissertation demonstrates the use of square-root and cube-root compression with a neuromorphic VLSI neuron to expand the range of distances over which ILD can be used to estimate echo direction in a sonar system based on echolocating bats. However, many questions remain regarding how to achieve the rapid control of a sonar-guided vehicle to pursue a goal while avoiding obstacles. Taking into account the limited field-of-view of practical sonar systems and vehicle kinematics, we propose an obstacle avoidance algorithm that maps the 2-D sensory space into a 1-D motor space and evaluates motor actions while combining obstacles and goal information. A winner-take-all (WTA) mechanism is used to select the final steering action. To avoid unnecessary scanning of the environment, an attentional system is proposed to control the directions of sonar pings for efficient, task-driven, sensory data collection. A mobile robot driven by the proposed algorithm was capable of navigating through a cluttered environment using a realistic sonar system. The algorithm was tested on a mobile robot, and it is implemented on a neural model using spike-timing representations, a spike-latency memory, and a “race-to-first-spike” WTA circuit. This dissertation also proposed a CMOS floating-gate circuit for artificial neural network synapse memories that can achieve a fixed rate of weight increase (adding vectors) or proportional decay (normalization) on the synaptic weight.Item Bio-Inspired Information Extraction In 3-D Environments Using Wide-Field Integration Of Optic Flow(2010) Hyslop, Andrew Maxwell; Humbert, James S; Aerospace Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)A control theoretic framework is introduced to analyze an information extraction approach from patterns of optic flow based on analogues to wide-field motion-sensitive interneurons in the insect visuomotor system. An algebraic model of optic flow is developed, based on a parameterization of simple 3-D environments. It is shown that estimates of proximity and speed, relative to these environments, can be extracted using weighted summations of the instantaneous patterns of optic flow. Small perturbation techniques are utilized to link weighting patterns to outputs, which are applied as feedback to facilitate stability augmentation and perform local obstacle avoidance and terrain following. Weighting patterns that provide direct linear mappings between the sensor array and actuator commands can be derived by casting the problem as a combined static state estimation and linear feedback control problem. Additive noise and environment uncertainties are incorporated into an offline procedure for determination of optimal weighting patterns. Several applications of the method are provided, with differing spatial measurement domains. Non-linear stability analysis and experimental demonstration is presented for a wheeled robot measuring optic flow in a planar ring. Local stability analysis and simulation is used to show robustness over a range of urban-like environments for a fixed-wing UAV measuring in orthogonal rings and a micro helicopter measuring over the full spherical viewing arena. Finally, the framework is used to analyze insect tangential cells with respect to the information they encode and to demonstrate how cell outputs can be appropriately amplified and combined to generate motor commands to achieve reflexive navigation behavior.Item GEOMETRIC COOPERATIVE CONTROL OF FORMATIONS(2004-11-11) Zhang, Fumin; Krishnaprasad, P.S.; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Robots in a team are modeled as particles which obey simple, second order dynamics. The whole team can be viewed as a deformable body with changing shape and orientation. Jacobi shape theory is applied to model such a formation. We derive the controlled system equations using the Lagrange-D'Alembert principle. Control forces on each robot are combined and reorganized as controls for the center, for rotation and for shape changes. From a shape-theoretic point of view, general feedback control laws are designed to achieve desired formations. The system equations on shape space provide possibilities for achieving formations without communication links between team members equipped with sufficient sensing ability. We allow each robot freedom to establish a coordinate system in which shape dynamics of the whole formation is computed. Without knowing such coordinate systems of other robots, each robot is able to perform cooperative control. This is made possible by a class of gauge covariant control laws. We argue that freedom of choosing gauge frame helps to improve controller performance. When all robots are required to have common constant speed, the control forces have to be of gyroscopic nature. Previous works of Justh and Krishnaprasad has inspired us to study the obstacle avoidance and navigation problem from a point of view of formation shape control. We achieve gyroscopic control laws to achieve boundary following behavior when the particle encounters an obstacle. The "steady state" trajectory of the particle forms a Bertrand pair with the boundary curve of the obstacle. This steady state behavior correspond to a relative equilibrium for a non-autonomous system on special Euclidean groups. Our control law achieves asymptotic convergence of the non-autonomous system dynamics. The boundary following behavior is a building block for robot navigation in a cluttered environment. Based on the configuration of the obstacles and the target, we may construct virtual boundary curves by analyzing sensory data. Such virtual boundary curves lead the robot to the target without collision. We have also studied the problem of establishing a formation of satellites with periodic shape changes near an elliptic earth orbit. We propose a control law that would set up a given formation near a given orbit. This law also allows a satellite formation to achieve orbit transfer.