AI is not a silver bullet, and no technology can replace aggressive political and corporate commitments to reducing emissions. But given the urgency, scale, and complexity of the global energy transition, we can't afford to leave any tools in the toolbox. Used well, AI will accelerate the energy transition while expanding access to energy services, encouraging innovation, and ensuring a safe, resilient, and affordable clean energy system. It is time for industry players and policy makers to lay the foundations for this AI-enabled energy future, and to build a trusted and collaborative ecosystem around AI for the energy transition.

Reinforcement learning has been found useful in solving optimal power flow (OPF) problems in electric power distribution systems. However, the use of largely model-free reinforcement learning algorithms that completely ignore the physics-based modeling of the power grid compromises the optimizer performance and poses scalability challenges. This paper proposes a novel approach to synergistically combine the physics-based models with learning-based algorithms using imitation learning to solve distribution-level OPF problems. Specifically, we propose imitation learning based improvements in deep reinforcement learning (DRL) methods to solve the OPF problem for a specific case of battery storage dispatch in the power distribution systems. The proposed imitation learning algorithm uses the approximate optimal solutions obtained from a linearized model-based OPF solver to provide a good initial policy for the DRL algorithms while improving the training efficiency. The effectiveness of the proposed approach is demonstrated using IEEE 34-bus and 123-bus distribution feeders with numerous distribution-level battery storage systems.

We study the problem of learning a range of vision-based manipulation tasks from a large offline dataset of robot interaction. In order to accomplish this, humans need easy and effective ways of specifying tasks to the robot. Goal images are one popular form of task specification, as they are already grounded in the robot's observation space. However, goal images also have a number of drawbacks: they are inconvenient for humans to provide, they can over-specify the desired behavior leading to a sparse reward signal, or under-specify task information in the case of non-goal reaching tasks. Natural language provides a convenient and flexible alternative for task specification, but comes with the challenge of grounding language in the robot's observation space. To scalably learn this grounding we propose to leverage offline robot datasets (including highly sub-optimal, autonomously collected data) with crowd-sourced natural language labels. With this data, we learn a simple classifier which predicts if a change in state completes a language instruction. This provides a language-conditioned reward function that can then be used for offline multi-task RL. In our experiments, we find that on language-conditioned manipulation tasks our approach outperforms both goal-image specifications and language conditioned imitation techniques by more than 25%, and is able to perform visuomotor tasks from natural language, such as "open the right drawer" and "move the stapler", on a Franka Emika Panda robot.

Crop type mapping at the field level is critical for a variety of applications in agricultural monitoring, and satellite imagery is becoming an increasingly abundant and useful raw input from which to create crop type maps. Still, in many regions crop type mapping with satellite data remains constrained by a scarcity of field-level crop labels for training supervised classification models. When training data is not available in one region, classifiers trained in similar regions can be transferred, but shifts in the distribution of crop types as well as transformations of the features between regions lead to reduced classification accuracy. We present a methodology that uses aggregate-level crop statistics to correct the classifier by accounting for these two types of shifts. To adjust for shifts in the crop type composition we present a scheme for properly reweighting the posterior probabilities of each class that are output by the classifier. To adjust for shifts in features we propose a method to estimate and remove linear shifts in the mean feature vector. We demonstrate that this methodology leads to substantial improvements in overall classification accuracy when using Linear Discriminant Analysis (LDA) to map crop types in Occitanie, France and in Western Province, Kenya. When using LDA as our base classifier, we found that in France our methodology led to percent reductions in misclassifications ranging from 2.8% to 42.2% (mean = 21.9%) over eleven different training departments, and in Kenya the percent reductions in misclassification were 6.6%, 28.4%, and 42.7% for three training regions. While our methodology was statistically motivated by the LDA classifier, it can be applied to any type of classifier. As an example, we demonstrate its successful application to improve a Random Forest classifier.

More than $3.5 billion in funding was funneled into 35 startups last month, much of that scattered across the globe. Several Chinese companies received significant funding as the country bulks up domestic production of wafers and GPUs. In addition, with attention increasing on the need for electric vehicles and renewable energy, big investments went into battery manufacturing startups. One company making EV batteries garnered $1.5 billion, while several other large rounds were targeted at grid-scale energy storage companies. Metax designs high-performance, reconfigurable GPUs based on its own instruction set for data center, gaming, and AI. Funds will be used for R&D, and the company recently launched a corporate research institute at Zhejiang University. Based in Shanghai, China, Metax was founded in 2020.

August 24, 2021 – Congratulations to Montreal-based start-up BrainBox AI, which is one of 10 companies--and the only Canadian company, it says--selected to present at COP26's "Tech for our planet" program, an initiative at the upcoming 26th United Nations Climate Change Conference. BrainBox was selected by the U.K. government to display its technology in Challenge 3–Thinking Smart, which is dedicated to solutions that can capture and share data to better predict and manage energy consumption. In the three months leading to COP26 (being held in November in Glasgow), BrainBox will demonstrate the benefits of grid-interactive buildings to achieve net zero objectives for the electrical grid. "By implementing technologies like BrainBox AI in one of the world's greatest energy consumers--buildings--we can turn the tide and help the real estate industry play its part in stopping the effects of climate change," said Sam Ramadori, president, BrainBox AI. Currently installed in over 100 million sf of real estate across 17 countries, BrainBox's flagship product combines AI and cloud computing to create a fully autonomous commercial HVAC solution that reduces energy consumption and emissions.

Complex Event Recognition (CER) systems have become popular in the past two decades due to their ability to "instantly" detect patterns on real-time streams of events. However, there is a lack of methods for forecasting when a pattern might occur before such an occurrence is actually detected by a CER engine. We present a formal framework that attempts to address the issue of Complex Event Forecasting (CEF). Our framework combines two formalisms: a) symbolic automata which are used to encode complex event patterns; and b) prediction suffix trees which can provide a succinct probabilistic description of an automaton's behavior. We compare our proposed approach against state-of-the-art methods and show its advantage in terms of accuracy and efficiency. In particular, prediction suffix trees, being variable-order Markov models, have the ability to capture long-term dependencies in a stream by remembering only those past sequences that are informative enough. Our experimental results demonstrate the benefits, in terms of accuracy, of being able to capture such long-term dependencies. This is achieved by increasing the order of our model beyond what is possible with full-order Markov models that need to perform an exhaustive enumeration of all possible past sequences of a given order. We also discuss extensively how CEF solutions should be best evaluated on the quality of their forecasts.

In this paper, we introduce a novel deep learning based solution to the Powered-Descent Guidance (PDG) problem, grounded in principles of nonlinear Stochastic Optimal Control (SOC) and Feynman-Kac theory. Our algorithm solves the PDG problem by framing it as an $\mathcal{L}^1$ SOC problem for minimum fuel consumption. Additionally, it can handle practically useful control constraints, nonlinear dynamics and enforces state constraints as soft-constraints. This is achieved by building off of recent work on deep Forward-Backward Stochastic Differential Equations (FBSDEs) and differentiable non-convex optimization neural-network layers based on stochastic search. In contrast to previous approaches, our algorithm does not require convexification of the constraints or linearization of the dynamics and is empirically shown to be robust to stochastic disturbances and the initial position of the spacecraft. After training offline, our controller can be activated once the spacecraft is within a pre-specified radius of the landing zone and at a pre-specified altitude i.e., the base of an inverted cone with the tip at the landing zone. We demonstrate empirically that our controller can successfully and safely land all trajectories initialized at the base of this cone while minimizing fuel consumption.

China is planning to build miles-wide'megastructures' in orbit, including solar power plants, tourism complexes, gas stations and even asteroid mining facilities. The National Natural Science Foundation of China (NSFC) announced a new five-year plan, directing researchers to develop technologies and techniques. The structures will require lightweight materials to allow larger objects to get into orbit with existing rockets. Researchers will also need to adopt technology to allow for in-orbit assembly and control. The Chinese government said there is an'urgent need' for megaprojects in space that would require ultra-large spacecraft to keep them in orbit.

For years, SimpliSafe's only option for outdoor video monitoring (other than the Video Doorbell Pro) was its indoor-only SimpliCam wrapped in a weatherproof rubber sleeve. Now, the company is finally offering a proper, battery-powered outdoor cam, complete with a weatherized shell, a spotlight, and people detection. Available now for $170, the SimpliSafe Wireless Outdoor Security Camera is a svelte, cylindrical camera with a swiveling magnetic base, Wi-Fi connectivity, Alexa and Google Assistant support, and a rechargeable, replaceable battery that promises to deliver between 3 and 6 months of battery life on a single charge. Equipped with two antennas, the new SimpliFi outdoor cam is capable of connecting to 2.4GHz Wi-Fi networks, but it won't work without a SimpliSafe base station. To get a base station, you'll need to purchase one of SimpliSafe's alarm systems, which start at $230 for a four-piece kit.