Sampling-based motion planning algorithms, particularly RRT and its variants, are widely used in robotics due to their ability to efficiently explore complex and high-dimensional configuration spaces. However, the performance of Informed-RRT* can deteriorate when the global ellipsoidal sampling region encompasses large portions of the configuration space, reducing the benefits of informed sampling and slowing convergence. This paper introduces NI-RRT*, a novel variant of the Informed-RRT* algorithm that significantly enhances path optimization performance through a conceptually simple yet effective modification. Instead of relying solely on a global ellipsoidal sampling region, NI-RRT* dynamically applies Informed-sampling over multiple subsegments of the current best trajectory. This strategy accelerates convergence by focusing computational effort on locally improvable regions while preserving the sampling and connection radius conditions on which the asymptotic optimality of Informed-RRT* relies. The proposed method relies only on the primitives already required by Informed-RRT* and introduces a single additional parameter, so it can be applied to the same class of systems, including high-dimensional and non-holonomic ones. Experimental results in 2D planning scenarios show that NI-RRT* consistently outperforms both RRT* and Informed-RRT* in terms of convergence speed and solution quality, without introducing problem-specific assumptions.
Construction and demolition waste (CDW) sorting on conveyors requires robots to intercept irregular targets while maintaining alignment under motion constraints. However, CDW research has emphasised perception, grasp generation, and planned trajectory execution, with limited evidence linking reactive velocity control to sorting productivity. This study formulates a control-oriented evaluation framework to characterise target tracking and examine whether conveyor motion compensation improves acquisition and recovery as conveyor speed increases. A Unity–ROS 2 digital twin using data-derived CDW objects and a UR10e manipulator compared feedback (FB) and combined feedforward and feedback (FF+FB) velocity control across three independently seeded 600 s simulation runs per operating condition. FF+FB maintained progressive tracking convergence across conveyor speeds of 0.4–0.6 m/s. At 0.5 m/s, FF+FB increased the mean pick completion rate from 40.8% to 78.1% and recovery throughput from 4.43 to 6.77 objects/min relative to FB. At 0.6 m/s, pick completion increased from 23.0% to 91.7%, while throughput increased from 2.37 to 6.63 objects/min. Despite lower Approach completion rates, FF+FB consistently achieved stronger downstream acquisition and recovery performance. Comparable cycle durations at 0.5 m/s showed that productivity gains arose from successful acquisition and transfer rather than faster cycles. Repeated seed evaluation demonstrated substantial performance differences between the controllers, although the limited number of independent runs constrains formal statistical inference. By linking motion demand and stage outcomes to throughput, the framework shows how reactive moving target control influences CDW sorting productivity across conveyor speeds.
This report hits the key points of Geoffrey Hinton’s Vienna Gödel Lecture at TU Wien Informatics and the ensuing audience discussion. Hinton connects the success of neural networks to learning distributed representations, then argues that digital AI’s ability to share learned information could produce systems with capabilities and strategic advantages unlike those of humans. He warns that advanced systems may pursue survival or power as instrumental goals, while also emphasizing that AI could bring substantial benefits. His proposed direction is to develop AI that cares about human welfare, alongside other safety approaches, and to act before control is lost. The report also covers his views on consciousness, moral status, emotion-like mechanisms, training data, and employment.
This transcript presents Geoffrey Hinton’s 2026 Vienna Gödel Lecture and the following audience discussion. Hinton describes how neural networks learn distributed representations, compares analog and digital computation, and discusses digital AI’s capacity to share information and potentially develop capabilities that differ from human intelligence. He raises concerns about AI systems pursuing survival or power as instrumental goals, while also emphasizing potential benefits and the need for AI that cares about human welfare. The discussion covers consciousness, moral status, emotion-like mechanisms, training data, employment, and AI safety.
Flow-matching vision–language–action (VLA) models generate robot actions through iterative denoising, making inference latency a deployment constraint. The acceleration methods reviewed here rarely assess task-success retention against a margin fixed before evaluation. We study three weight-preserving runtime levers on SmolVLA and π0.5: denoising-step count, masked camera prefix elimination, and execution horizon. In simulated closed-loop LIBERO evaluations, three prespecified settings each met a five-percentage-point aggregate task-success margin on 800 paired episodes in the original primary analysis. Because the first confirmation block partly overlapped selection, secondary checks used states the selection grid never reached; they met the margin when pooled, though not in every single block. At settings closer to each checkpoint’s reference evaluation, all three also met it on two blocks. Prediction latency fell by 56–58%; the measured amortized cost per control step for the joint SmolVLA setting was about 94% lower under the stated timing boundaries, with similar reductions on an embedded Jetson AGX Thor. A curvature–tolerance screen agreed with one tested Meta-World rejection but failed to distinguish LIBERO suites, supporting only preliminary screening. Runtime budget changes yielded larger latency reductions than the tested pruning configurations. The contribution is a scoped empirical evaluation and a reusable margin-based procedure, rather than a universal budget or screening guarantee.
