Self-localization is an important technology for automating bulldozers. Conventional bulldozer self-localization systems rely on RTK-GNSS (Real Time Kinematic-Global Navigation Satellite Systems). However, RTK-GNSS signals are sometimes lost in certain mining conditions. Therefore, self-localization methods that do not depend on RTK-GNSS are required. In this paper, we propose a machine learning-based self-localization method for bulldozers. The proposed method consists of two steps: estimating local velocities using a machine learning model from internal sensors, and incorporating these estimates into an Extended Kalman Filter (EKF) for global localization. We also created a novel dataset for bulldozer odometry and conducted experiments across various driving scenarios, including slalom, excavation, and driving on slopes. The result demonstrated that the proposed self-localization method suppressed the accumulation of position errors compared to kinematics-based methods, especially when slip occurred. Furthermore, this study showed that bulldozer-specific sensors, such as blade position sensors and hydraulic pressure sensors, contributed to improving self-localization accuracy.

Israel killed three Palestinians in a drone strike on Thursday in Tulkarem, a city and refugee camp in the occupied West Bank. That was during an Israeli raid – a near-daily occurrence in the West Bank – on the Tulkarem refugee camp, during which Israeli troops clashed with fighters from the Qassam Brigades, the military wing of Hamas, according to fighters in the city. Here's all you need to know about Israeli raids on Tulkarem: News reports say Israeli soldiers were deployed on rooftops and sent bulldozers into the camp to destroy large residential areas. Israel also reportedly set fire to people's homes and prevented local relief workers from putting the fires out. Experts say Israel's tactics during its raids appear to be part of a broader doctrine to collectively punish the population, ostensibly because pockets of armed resistance are fighting back against Israel's ever-entrenching occupation. Israel claims that it is conducting "counter-terrorism" operations.

Text-driven localized editing of 3D objects is particularly difficult as locally mixing the original 3D object with the intended new object and style effects without distorting the object's form is not a straightforward process. To address this issue, we propose a novel NeRF-based model, Blending-NeRF, which consists of two NeRF networks: pretrained NeRF and editable NeRF. Additionally, we introduce new blending operations that allow Blending-NeRF to properly edit target regions which are localized by text. By using a pretrained vision-language aligned model, CLIP, we guide Blending-NeRF to add new objects with varying colors and densities, modify textures, and remove parts of the original object. Our extensive experiments demonstrate that Blending-NeRF produces naturally and locally edited 3D objects from various text prompts. Our project page is available at https://seokhunchoi.github.io/Blending-NeRF/

Granular materials are of critical interest to many robotic tasks in planetary science, construction, and manufacturing. However, the dynamics of granular materials are complex and often computationally very expensive to simulate. We propose a set of methodologies and a system for the fast simulation of granular materials on Graphics Processing Units (GPUs), and show that this simulation is fast enough for basic training with Reinforcement Learning algorithms, which currently require many dynamics samples to achieve acceptable performance. Our method models granular material dynamics using implicit timestepping methods for multibody rigid contacts, as well as algorithmic techniques for efficient parallel collision detection between pairs of particles and between particle and arbitrarily shaped rigid bodies, and programming techniques for minimizing warp divergence on Single-Instruction, Multiple-Thread (SIMT) chip architectures. We showcase our simulation system on several environments targeted toward robotic tasks, and release our simulator as an open-source tool.

Most people think that AI is competing against humans. They argue that It's dangerous to humans and also it's here to snatch their livelihood. They have a picture in their mind of robots ruling over humanity in the year 2050 or robots doing standard industry work while humans are starving as they are unemployed. Clearly, a wild dystopian idea cultivated by the movies and some anti-AI "thinkers". Similar arguments were given when electricity and heavy machines like bulldozers were invented. People feared that electricity will kill humans/cattle and that machines like bulldozers will strip them of their livelihood.

In this work, we aim to tackle the problem of autonomous grading, where a dozer is required to flatten an uneven area. In addition, we explore methods for bridging the gap between a simulated environment and real scenarios. We design both a realistic physical simulation and a scaled real prototype environment mimicking the real dozer dynamics and sensory information. We establish heuristics and learning strategies in order to solve the problem. Through extensive experimentation, we show that although heuristics are capable of tackling the problem in a clean and noise-free simulated environment, they fail catastrophically when facing real world scenarios. As the heuristics are capable of successfully solving the task in the simulated environment, we show they can be leveraged to guide a learning agent which can generalize and solve the task both in simulation and in a scaled prototype environment.

As a teenager working for his dad's construction business, Noah Ready-Campbell dreamed that robots could take over the dirty, tedious parts of his job, such as digging and leveling soil for building projects. Now the former Google engineer is turning that dream into a reality with Built Robotics, a startup that's developing technology to allow bulldozers, excavators and other construction vehicles to operate themselves. "The idea behind Built Robotics is to use automation technology make construction safer, faster and cheaper," said Ready-Campbell, standing in a dirt lot where a small bulldozer moved mounds of earth without a human operator. The San Francisco startup is part of a wave of automation that's transforming the construction industry, which has lagged behind other sectors in technological innovation. Backed by venture capital, tech startups are developing robots, drones, software and other technologies to help the construction industry to boost speed, safety and productivity.

As a teenager working for his dad's construction business, Noah Ready-Campbell dreamed that robots could take over the dirty, tedious parts of his job, such as digging and leveling soil for building projects. Now the former Google engineer is turning that dream into a reality with Built Robotics, a startup that's developing technology to allow bulldozers, excavators and other construction vehicles to operate themselves. 'The idea behind Built Robotics is to use automation technology make construction safer, faster and cheaper,' said Ready-Campbell, standing in a dirt lot where a small bulldozer moved mounds of earth without a human operator. The San Francisco startup is part of a wave of automation that's transforming the construction industry, which has lagged behind other sectors in technological innovation. Backed by venture capital, tech startups are developing robots, drones, software and other technologies to help the construction industry to boost speed, safety and productivity.

Robots have moved into factories, warehouses, stores and even our homes. Tech startups are developing self-driving bulldozers, drones to inspect work sites and robot bricklayers. In this photo taken Jan. 26, 2018, Mike Moy, an assistant plant manager for Lehigh Hanson Cement Group, inspects a Kespry drone he uses to survey inventories of rock, sand and other building materials at a mining plant in Sunol, California. Robots are coming to a construction site near you. Tech startups are developing self-driving bulldozers, survey drones and bricklaying robots to help the construction industry boost productivity, speed and safety as it struggles to find enough skilled workers.

This figure shows a small fraction (about 7 km by 8 km) of the entire 75-km-square map. The northern tip of Yellowstone Lake is at the bottom of the screen. Thin black lines represent elevation contours, slightly wider lines represent roads, and the widest lines represent the fireline cut by bulldozers. Blue lines represent rivers and streams. The blue B in the bottom left corner marks the location of the fireboss, the agent that directs all the others.