Lifeboat Foundation

Designed for precision agriculture and environmental management use cases, the P4 Multispectral drone combines data from six separate sensors to measure the health of crops. It can be used to monitor everything from individual plants to entire fields, as well as weeds, insects, and a variety of soil conditions.

The P4 Multispectral drone is compatible with standard industry workflows including flight programming, mapping, and analytics software from DJI and other leading providers. Using the DJI GS Pro application, you can create automated and repeatable missions including flight planning, mission execution, and flight data management. Data collected can be easily imported into DJI Terra or a suite of third-party software including Pix4D Mapper and DroneDeploy, for analysis and to generate additional vegetation index maps.

Continue reading “DJI releases new firmware update for Phantom 4 Multispectral drone” »

Normally, robotic arms are controlled by a GUI running on a host PC, or with some kind of analog system that maps human inputs to various degrees of rotation. However, Maurizio Miscio was able to build a custom robotic arm that is completely self-contained — thanks to a companion mobile app that resides on an old smartphone housed inside a control box.

Miscio started his project by making 3D models of each piece, most of which were 3D-printed. These included the gripper, various joints that each give a single axis of rotation, and a large circular base that acts as a stable platform on which the arm can spin. He then set to work attaching five servo motors onto each rotational axis, along with a single SG90 micro servo motor for the gripper. These motors were connected to an Arduino Uno that also had an HC-05 Bluetooth® serial module for external communication.

In order to operate the arm, Miscio developed a mobile app with the help of MIT App Inventor, which presents the user with a series of buttons that rotate a particular servo motor to the desired degree. The app even lets a series of motion be recorded and “played back” to the Uno over Bluetooth for repeated, accurate movements.

By Planning in the Latent Space of a Learned World Model. The world model Director builds from pixels allows effective planning in a latent space. To anticipate future model states given future actions, the world model first maps pictures to model states. Director optimizes two policies based on the model states’ anticipated trajectories: Every predetermined number of steps, the management selects a new objective, and the employee learns to accomplish the goals using simple activities. The direction would have a difficult control challenge if they had to choose plans directly in the high-dimensional continuous representation space of the world model. To reduce the size of the discrete codes created by the model states, they instead learn a goal autoencoder. The goal autoencoder then transforms the discrete codes into model states and passes them as goals to the worker after the manager has chosen them.

Deep reinforcement learning advancements have accelerated the study of decision-making in artificial agents. Artificial agents may actively affect their environment by moving a robot arm based on camera inputs or clicking a button in a web browser, in contrast to generative ML models like GPT-3 and Imagen. Although artificial intelligence has the potential to aid humans more and more, existing approaches are limited by the necessity for precise feedback in the form of often given rewards to acquire effective techniques. For instance, even robust computers like AlphaGo are restricted to a certain number of moves before earning their next reward while having access to massive computing resources.

Contrarily, complex activities like preparing a meal necessitate decision-making at all levels, from menu planning to following directions to the shop to buy supplies to properly executing the fine motor skills required at each stage along the way based on high-dimensional sensory inputs. Artificial agents can complete tasks more independently with scarce incentives thanks to hierarchical reinforcement learning (HRL), which automatically breaks down complicated tasks into achievable subgoals. Research on HRL has, however, been difficult because there is no universal answer, and existing approaches rely on manually defined target spaces or subtasks.

Astronomers from MIT report today that they have discovered a mysterious signal with a pattern akin to a heartbeat, emanating from a far-off galaxy that is billions of light-years from Earth. Exactly what the source may be of this regular pulse of radio waves remains a mystery, as it is the first time that such a signal has been recorded.

They have identified the signal as a fast radio burst (FRB), which is typically an intensely strong burst of radio waves of unknown astrophysical origin that lasts only a few milliseconds at most. This new signal, labelled FRB 20191221A, is unusual, because it persists for up to three seconds, which is about 1,000 times longer than the average FRB. Within this time, there are shorter bursts of radio waves that repeat every 0.2 seconds in a clear periodic pattern, similar to that of a beating heart.

Since the first FRB was discovered in 2007, hundreds of similar radio flashes have been detected across the universe, most recently by the Canadian Hydrogen Intensity Mapping Experiment, or CHIME, an interferometric radio telescope that is located at the Dominion Radio Astrophysical Observatory in British Columbia, Canada. CHIME is designed to pick up radio waves emitted by hydrogen in the very earliest stages of the universe, but the telescope is also sensitive to fast radio bursts. Since it began observing the sky in 2018, CHIME has detected hundreds of FRBs emanating from different parts of the sky.

Powerful radio pulses originating deep in the cosmos can be used to study hidden pools of gas cocooning nearby galaxies, according to a new study appearing in the journal Nature Astronomy.

So-called fast radio bursts, or FRBs, are pulses of radio waves that typically originate millions to billions of light-years away (radio waves are electromagnetic radiation like the light we see with our eyes but have longer wavelengths and frequencies). The first FRB was discovered in 2007, and since then, hundreds more have been found. In 2020, Caltech’s STARE2 instrument (Survey for Transient Astronomical Radio Emission 2) and Canada’s CHIME (Canadian Hydrogen Intensity Mapping Experiment) detected a massive FRB that went off in our own Milky Way galaxy. Those earlier results helped confirm the theory that the energetic events most likely originate from dead, magnetized stars called magnetars.

As more and more FRBs roll in, researchers are now asking how they can be used to study the gas that lies between us and the bursts. In particular, they would like to use the FRBs to probe halos of diffuse gas that surround galaxies. As the radio pulses travel toward Earth, the gas enveloping the galaxies is expected to slow the waves down and disperse the radio frequencies. In the new study, the researchers looked at a sample of 474 distant FRBs detected by CHIME, which has discovered the most FRBs to date, and showed that the subset of two dozen FRBs that passed through galactic halos were indeed slowed down more than non-intersecting FRBs.

On-chip training of machine learning algorithms is challenging for photonic devices. Here, the authors construct nonlinear mapping functions in silicon photonic circuits, and experimentally demonstrate on-chip bacterial foraging training for projection-based classification.

It’s effectively a new data set that will fuel the second wave of discoveries about Mars’ surface composition.

But while it was doing that work, it was also gathering lower-resolution mapping strips, about 83,000 of them. Now that CRISM is no longer active, the team is building their map from those strips.

Processing this much data into one cohesive map is a complicated task requiring powerful computing resources. It takes time to optimize the maps and account for environmental conditions and discrepancies between the different images.

Continue reading “NASA just built the best map of Mars to date using 51,000 images” »

We know the brain changes after traumatic injury, and now we have maps from mice showing what that change looks like.

A team of scientists has traced connections between nerve cells throughout the entire brain of mice, showing that distant parts of the brain become disconnected after a head injury.

The stunning visualizations of brain-wide connectivity could help scientists understand how a traumatic brain injury, or TBI, alters cross-talk between different cells and brain regions, first in mice and then in humans.

An Interview with COO Dijam Panigrahi.

“a unified and shared software infrastructure to empower enterprise customers to build and run scalable, high-quality eXtended Reality (XR) – Augmented Reality (AR), Virtual Reality (VR) and Mixed Reality (MR) – applications in public, private, and hybrid clouds.”

What does that all mean?

Continue reading “GridRaster Uses the Metaverse to Build High Tech Prototypes and Finished Products” »