Robustness of Bio-Inspired Visual Systems for Collision Prediction in Critical Robot Traffic

Collision prevention sets a major research and development obstacle for intelligent robots and vehicles. This paper investigates the robustness of two state-of-the-art neural network models inspired by the locust’s LGMD-1 and LGMD-2 visual pathways as fast and low-energy collision alert systems in critical scenarios. Although both the neural circuits have been studied and modelled intensively, their capability and robustness against real-time critical traffic scenarios where real-physical crashes will happen have never been systematically investigated due to difficulty and high price in replicating risky traffic with many crash occurrences. To close this gap, we apply a recently published robotic platform to test the LGMDs inspired visual systems in physical implementation of critical traffic scenarios at low cost and high flexibility. The proposed visual systems are applied as the only collision sensing modality in each micro-mobile robot to conduct avoidance by abrupt braking. The simulated traffic resembles on-road sections including the intersection and highway scenes wherein the roadmaps are rendered by coloured, artificial pheromones upon a wide LCD screen acting as the ground of an arena. The robots with light sensors at bottom can recognise the lanes and signals, tightly follow paths. The emphasis herein is laid on corroborating the robustness of LGMDs neural systems model in different dynamic robot scenes to timely alert potential crashes. This study well complements previous experimentation on such bio-inspired computations for collision prediction in more critical physical scenarios, and for the first time demonstrates the robustness of LGMDs inspired visual systems in critical traffic towards a reliable collision alert system under constrained computation power. This paper also exhibits a novel, tractable, and affordable robotic approach to evaluate online visual systems in dynamic scenes.

CT-Based Collision Prediction Software for External-Beam Radiation Therapy

PurposeBeam angle optimization is a critical issue for modern radiotherapy (RT) and is a challenging task, especially for large body sizes and noncoplanar designs. Noncoplanar RT techniques may have dosimetric advantages but increase the risk of mechanical collision. We propose a software solution to accurately predict colliding/noncolliding configurations for coplanar and noncoplanar beams.Materials and MethodsIndividualized software models for two different linear accelerators were built to simulate noncolliding gantry orientations for phantom/patient subjects. The sizes and shapes of the accelerators were delineated based on their manuals and on-site measurements. The external surfaces of the subjects were automatically contoured based on computed tomography (CT) simulations. An Alderson Radiation Therapy phantom was used to predict the accuracy of spatial collision prediction by the software. A gantry collision problem encountered by one patient during initial setup was also used to test the validity of the software. Results: In the comparison between the software estimates and on-site measurements, the noncoplanar collision angles were all predicted within a 5-degree difference in gantry position. The confusion matrix was calculated for each of the two empty accelerator models, and the accuracies were 98.7% and 97.3%. The true positive rates were 97.7% and 96.9%, while the true negative rates were 99.8% and 97.9%, respectively. For the phantom study, the collision angles were predicted within a 5-degree difference. The software successfully predicted the collision problem encountered by the breast cancer patient in the initial setup position and generated shifted coordinates that were validated to correspond to a noncolliding geometry.ConclusionThe developed software effectively and accurately predicted collisions for accelerator-only, phantom, and patient setups. This software may help prevent collisions and expand the range of spatially applicable beam angles.