Inferring physical properties can significantly enhance robotic manipulation by enabling robots to handle objects safely and efficiently through adaptive grasping strategies. Previous approaches have typically relied on either tactile or visual data, limiting their ability to fully capture properties. We introduce a novel cross-modal perception framework that integrates visual observations with tactile representations within a multimodal vision-language model. Our physical reasoning framework, which employs a hierarchical feature alignment mechanism and a refined prompting strategy, enables our model to make property-specific predictions that strongly correlate with ground-truth measurements. Evaluated on 35 diverse objects, our approach outperforms existing baselines and demonstrates strong zero-shot generalization.

Perception systems in robotics encounter significant challenges in high-speed and dynamic conditions when relying on traditional cameras, where motion blur can compromise spatial feature integrity and task performance. Brain-inspired vision sensors (BVS) have recently gained attention as an alternative, offering high temporal resolution with reduced bandwidth and power requirements. Here, we present the first quantitative evaluation framework for two representative classes of BVSs in variable-speed robotic sensing, including event-based vision sensors (EVS) that detect asynchronous temporal contrasts, and the primitive-based sensor Tianmouc that employs a complementary mechanism to encode both spatiotemporal changes and intensity. A unified testing protocol is established, including crosssensor calibrations, standardized testing platforms, and quality metrics to address differences in data modality. From an imaging standpoint, we evaluate the effects of sensor non-idealities, such as motion-induced distortion, on the capture of structural information. For functional benchmarking, we examine task performance in corner detection and motion estimation under different rotational speeds. Results indicate that EVS performs well in highspeed, sparse scenarios and in modestly fast, complex scenes, but exhibits performance limitations in high-speed, cluttered settings due to pixel-level bandwidth variations and event rate saturation. In comparison, Tianmouc demonstrates consistent performance across sparse and complex scenarios at various speeds, supported by its global, precise, high-speed spatiotemporal gradient samplings. These findings offer valuable insights into the applicationdependent suitability of BVS technologies and support further advancement in this area.

Robotic perception refers to the ability of robots to sense and understand their environment. This is a critical component of artificial intelligence and plays a key role in enabling robots to perform tasks in a variety of settings. Whether it's detecting obstacles, recognizing objects, or locating landmarks, the ability to perceive and understand the environment is essential for robots to function effectively. One of the main challenges of robotic perception is that the environment can be highly dynamic and uncertain. Robots must be able to adapt to new situations, deal with changing conditions, and overcome unexpected obstacles.

To better optimise the global food supply chain, robotic solutions are needed to automate tasks currently completed by humans. Namely, phenotyping, quality analysis and harvesting are all open problems in the field of agricultural robotics. Robotic perception is a key challenge for autonomous solutions to such problems as scene understanding and object detection are vital prerequisites to any grasping tasks that a robot may undertake. This work conducts a brief review of modern robot perception models and discusses their efficacy within the agri-food domain.

Britannica defines a robot as any automatically operated machine that replaces human effort, though it may not resemble human beings in appearance or perform functions in a humanlike manner. By extension, robotics is the engineering discipline dealing with robots' design, construction, and operation. Robots have increasingly been used in environments that require object grasping and manipulation. They are helpful in households where there may be a need to pick and place objects such as books, balls, toys, and manufacturing productions lines to pick and move products such as packaged goods and mechanical parts. Research in robotic grasping and manipulation is believed to have started as far back as the 1970s, mainly when science-fiction classic Westworld staged robots in a fictitious film.

Robotics perception research has advanced tremendously in recent years thanks to the development of affordable and cutting edge sensor technologies (e.g., LiDAR, sonars, etc.) and data-driven techniques. While progress is still being made, several of these methods are trained, applied and evaluated with abundant and high-quality data. However, many field or in-the-wild robotics applications face substantial performance drops with respect to applications in constrained/structured environments due to low-quality visual data common in these scenarios; which suffers from various types of degradation and environmental disturbances (fog, ash, or inclement weather). And although some of these artifacts can be overcome by sophisticated algorithms and models, their impact becomes more noticeable as the level of degradation or change passes some empirical threshold. Based on this, and as an extension of the ICRA 2019 workshop on "Underwater Robotics Perception", the goal of this Research Topic is to review the recent progress of robust visual perception technologies and methods in challenging adversarial environments.

For self-driving cars and other smart transport to be successfully integrated in the real-world, the safety of passengers and pedestrians must be ensured. In the world of intelligent machines, perception answers the question: what is around me? This situational awareness is paramount for safe operation of autonomous vehicles in real-world environments. Scientists working in this field point to robotic perception as fundamental in equipping machines with a semantic understanding of the world, so that they can reliably identify objects and make informed predictions and actions. Michael Milford, Associate Professor at Queensland University of Technology (QUT), is a leading robotics researcher working to improve perception and more in autonomous vehicles, conducting his research at the intersection of robotics, neuroscience and computer vision.