No abstract

Robust and efficient target-tracking algorithms embedded on moving platforms, are a requirement for many computer vision and robotic applications. However, deployment of a real-time system is challenging, even with the computational power of modern hardware. As inspiration, we look to biological lightweight solutions-lightweight and low-powered flying insects. For example, dragonflies pursue prey and mates within cluttered, natural environments, deftly selecting their target amidst swarms. In our laboratory, we study the physiology and morphology of dragonfly 'small target motion detector' neurons likely to underlie this pursuit behaviour. Here we describe our insect-inspired tracking model derived from these data and compare its efficacy and efficiency with state-of-the-art engineering models. For model inputs, we use both publicly available video sequences, as well as our own task-specific dataset (small targets embedded within natural scenes). In the context of the tracking problem, we describe differences in object statistics within the video sequences. For the general dataset, our model often locks on to small components of larger objects, tracking these moving features. When input imagery includes small moving targets, for which our highly nonlinear filtering is matched, the robustness outperforms state-of-the-art trackers. In all scenarios, our insect-inspired tracker runs at least twice the speed of the comparison algorithms.

Although flying insects have limited visual acuity (approx. 18) and relatively small brains, many species pursue tiny targets against cluttered backgrounds with high success. Our previous computational model, inspired by electrophysiological recordings from insect 'small target motion detector' (STMD) neurons, did not account for several key properties described from the biological system. These include the recent observations of response 'facilitation' (a slow build-up of response to targets that move on long, continuous trajectories) and 'selective attention', a competitive mechanism that selects one target from alternatives. Here, we present an elaborated STMD-inspired model, implemented in a closed loop target-tracking system that uses an active saccadic gaze fixation strategy inspired by insect pursuit. We test this system against heavily cluttered natural scenes. Inclusion of facilitation not only substantially improves success for even short-duration pursuits, but it also enhances the ability to 'attend' to one target in the presence of distracters. Our model predicts optimal facilitation parameters that are static in space and dynamic in time, changing with respect to the amount of background clutter and the intended purpose of the pursuit. Our results provide insights into insect neurophysiology and show the potential of this algorithm for implementation in artificial visual systems and robotic applications.

Robot frame compliance has a large negative effect on the global accuracy of the system when large external forces/torques are exerted. This phenomenon is particularly problematic in applications where the robot is required to achieve ultrahigh (micron level) accuracy under very large external loads, e.g., in biomechanical testing and high precision machining. To ensure the positioning accuracy of the robot in these applications, the authors proposed a novel Stewart platform-based manipulator with decoupled sensor–actuator locations. The unique mechanism has the sensor locations fully decoupled from the actuator locations for the purpose of passively compensating for the load frame compliance, as a result improving the effective stiffness of the manipulator in six degrees of freedom (6DOF). In this paper, the stiffness of the proposed manipulator is quantified via a simplified method, which combines both an analytical model (robot kinematics error model) and a numerical model [finite element analysis (FEA) model] in the analysis. This method can be used to design systems with specific stiffness requirements. In the control aspect, the noncollocated positions of the sensors and actuators lead to a suboptimal control structure, which is addressed in the paper using a simple Jacobian-based decoupling method under both kinematics- and dynamics-based control. Simulation results demonstrate that the proposed manipulator configuration has an effective stiffness that is increased by a factor of greater than 15 compared to a general design. Experimental results show that the Jacobian-based decoupling method effectively increases the dynamic tracking performance of the manipulator by 25% on average over a conventional method.

Our results provide insight into the neuronal mechanisms that underlie biological target detection and selection (from a moving platform), as well as highlight the effectiveness of our bio-inspired algorithm in an artificial visual system.