How does one deal with the unexpected? Our world is full of surprises and we humans are often able to correctly identify a problem and respond appropriately. Consider a new driver encountering their first traffic circle; a student experiencing a hard drive failure in the middle of an assignment; an unexpected question being asked during a job interview. In situations where we have a goal (i.e., reach a destination or submit a completed assignment), we may need to alter our original plan when the unexpected occurs. Could we enable autonomous artificial intelligent agents to do the same?
A long standing area of artificial intelligence is the field of automated planning. The traditional planning problem is to generate a sequence of actions given a concrete, specific goal (e.g., I will be home at dinnertime) and a set of specific actions (e.g., drive-car, fill-gas-tank, walk, etc). Generating plans that are hopefully efficient and optimal from start to finish under different circumstances (e.g., delayed effects) is an active area of research. After a plan has been generated, and during the execution of the plan, the environment may change. For example, a robot retrieving packages in a warehouse may discover it has dropped its package. Or perhaps another robot has broken down due to a hardware failure and is blocking the path of this robot. How can a robot (or any A.I. agent) know something unexpected has happened without knowing all possible future failures?
Fundamental research on autonomy aims to find general approaches to solve this problem. One approach is to generate expectations: facts that should be true during different stages of a plan's execution. When an expectation is violated, a discrepancy occurs between the expected and perceived facts. A new trend in autonomy is to include goal reasoning capabilities. In the event of a failure, the original goal may no longer be warranted. Perhaps robust autonomous agents need to generate and change their goals in response to a changing environment.
Autonomous systems still have a long way to go and open research questions on autonomous systems remain. Funding agencies consistently seek new research on autonomy for diverse operations ranging from cybersecurity to military and vehicular autonomy. What will autonomous systems be like in the future? Will we achieve autonomous agents that can handle any situation they encounter?
- Dustin Dannenhauer
In this issue, we are pleased to feature expert commentary by Eric Sandosham, Founder and Partner at Red & White Consulting Partners LLP. Formerly Managing Director and Regional Head of Citibank's Decision Management and Analytics function in 14 global markets, Eric is currently based in Singapore. ...
We recently witnessed one of the biggest game AI events in history – Alpha Go became the first computer program to beat the world champion in a game of Go. The publication can be found here. Different techniques from machine learning and tree search have been combined by developers from DeepMind to ...
A recent press release states, "ClickSoftware, the leading provider of field service management software, today announced significant new capabilities for Field Service Edge, its cloud-based, mobile workforce management platform designed to meet the needs of the most demanding field service organizations. This latest offering introduces new strategic capabilities that will greatly improve field service efficiency and effectiveness. Major features are: (1) Predictive field service powered by ClickSoftware's Machine Learning Cloud, which identifies data patterns to make predictions and automatically improve valuable KPIs, (2) New demand forecasting capabilities to support more accurate resource planning and schedule optimization and provide richer insights to support proper staffing.
An AirAsia flight AK416 from Kuala Lumpur to Bandung, capital of Indonesia's West Java province was diverted to Johor's Senai International Airport on Wednesday after a male employee died mid-flight. The budget airlines confirmed the news in a report by the New Straits Times that the flight was diverted to Johor, a state in southern Malaysia because of a "medical emergency". In this photo, an Indian airport staff member walks next to an AirAsia airplane after it landed on its inaugural flight from New Delhi to Bagdogra Airport on Feb. 19, 2017. The airlines stated that although the staff member, who remained unidentified, was given immediate medical attention upon landing, he was later pronounced dead by a doctor on the ground. "The AirAsia family is saddened by the loss of an All-star and extends our deepest sympathies to the family of the deceased.
The concept of artificial intelligence (AI), described as the development of computer systems capable of performing tasks which, in general, require human intelligence, arose more than 60 years ago.However, this technology did not reach the field of financial services until the early 1980s.It had limited implementation and use at that time due to the state of the technology and power of computer systems. But in recent years there has been significant growth in injecting AI into the financial planning process. In the early 1980s, for example, the Citibank Investment Bank attempted to build expert systems, using artificial intelligence that imitated the decision-makingpower of a human expert. And Citibank was not the only one, many other Wall Street companies launched similar projects at that time. And in 1987, the Security Pacific National Bank launched a Fraud Prevention Task Force to automatically counter, through the use of artificial intelligence, unauthorized use of debit cards at ATMs and stores.
The subject of game AI generally begins with so-called perfect information games. These are turn-based games where the players have no information hidden from each other and there is no element of chance in the game mechanics (such as by rolling dice or drawing cards from a shuffled deck). Tic Tac Toe, Connect 4, Checkers, Reversi, Chess, and Go are all games of this type. Because everything in this type of game is fully determined, a tree can, in theory, be constructed that contains all possible outcomes, and a value assigned corresponding to a win or a loss for one of the players. Finding the best possible play, then, is a matter of doing a search on the tree, with the method of choice at each level alternating between picking the maximum value and picking the minimum value, matching the different players' conflicting goals, as the search proceeds down the tree.
September through December are the busiest cargo shipping months of the year thanks to the winter holiday season, and in 2017, that was even more true than usual. The demand for shipping space on container ships, and the pace of arrivals at commercial ports, can hit companies with time-consuming and expensive issues: shipment delays, required changes in shipping method from marine to air, scheduling problems for the unloading and reloading of containers, and freight theft. In a retail environment where Amazon and other large retailers offer quick shipping, for free, manufacturers and retailers now risk losing money -- and customers -- if deliveries are delayed. Increasingly, the commercial shipping firms that retailers and manufacturers rely on to get products from A to B are turning to new technologies like artificial intelligence and automation to analyze the huge amounts of data generating in shipping, with an eye toward streamlining the processes, anticipating potential delays, and saving money. For an industry that has used some of the same systems for years, artificial intelligence and automation offer an opportunity for revolution.
Drew McDermott Research on planning for robots is in such a state of flux that there is disagreement about what planning is and whether it is necessary. We can take planning to be the optimization and debugging of a robot's program by reasoning about possible courses of execution. It is necessary to the extent that fragments of robot programs are combined at run time. There are several strands of research in the field; I survey six: (1) attempts to avoid planning; (2) the design of flexible plan notations; (3) theories of time-constrained planning; (4) planning by projecting and repairing faulty plans; (5) motion planning; and (6) the learning of optimal behaviors from reinforcements. More research is needed on formal semantics for robot plans.
To date, using Tesla's trip planning tool has meant sitting inside your electric car while you map a route that takes you past charging stations. That doesn't make much sense if you're gearing up for vacation, does it? There's now a better way: Tesla has launched a web version of its trip planner to use while you're still sitting at your desk. It's not as fleshed out as the in-car version, but it can tell you where you'll need to charge and how long you need to drive based on both the route and the particular Tesla you're driving. You could see fewer stops with a Model S P100D than you would with a Model X 75D, for instance.