\n
![\"DTLSectors\"](\"\/msl\/files\/2014\/01\/DTLSectors.png\")
<\/a><\/p>\n<\/p>\n
In this scenario, the aerial robot is tasked with locating human survivors and potential fires in an urban area immediately after a natural disaster, such as an earthquake. \u00a0If \u00a0a survivor is found in an unsafe region (denoted by red), then the robot must guide the humans to a safe pickup region (green) before reporting its location at a data upload hub (blue). \u00a0If the survivor is in a safe region or a fire is found, then the robot needs to report its location at a data upload region. \u00a0Once the robot has localized survivors as well as possible, it must exit the scene (orange regions).<\/p>\n
In order to complete this task, the robot must plan its motion on-line such that it gathers information about some spatial feature (survivor locations), reacts to gains in information (report locations of survivors when they are found) and organize tasks sequentially (ensure survivor safety before reporting locations). \u00a0 While state-of-the-art active sensing algorithms deal very well with information gathering, they are not easily extended to the other planning tasks.<\/p>\n
In order to integrate reactive and sequential planning into active sensing, we turn to the field of temporal logic planning. \u00a0Temporal logics \u00a0(TLs) are extensions of Boolean logic that integrate how the state of a system may evolve over time . \u00a0An example of a task that may be described by a TL formula is “Visit Region A and then Region B while always avoiding hazards. \u00a0If Region C is entered, go to region D before entering region B. ” \u00a0Such rich missions cannot be decomposed into simple “go from point A to point B” specifications \u00a0We combine TLs with active sensing in the following two ways.<\/p>\n
Temporal Logic-Constrained Informative Path Planning<\/strong><\/p>\n We first consider applications in which the temporal logic mission is given\u00a0over known features of the environment, e.g. the robot is operating in an\u00a0environment with known topology and has to avoid hazards while visiting a\u00a0sequence of regions. The robot is also tasked with forming a minimal\u00a0uncertainty (minimal Shannon entropy) measure of some a priori unknown feature,\u00a0a problem known as the informative path planning problem.<\/p>\n Preliminary results may be found here<\/a>. We have developed a way to combine TL\u00a0and informative planning into a single mathematical framework and developed two\u00a0algorithms, an off-line expectation maximization and an on-line receding\u00a0horizon procedure, to solve this problem. We have developed stochastic\u00a0dynamic programming versions of these algorithms. Currently, we are validating\u00a0these results on experimental platforms. Future research directions include\u00a0extending our results to multi-agent systems.<\/p>\n Distribution Temporal Logic<\/strong><\/p>\n Next, we consider applications in which the temporal logic mission is given\u00a0over unknown features of the environment, e.g. in the above example, the robot\u00a0must find the survivor and guide it to a safe location. In order to address\u00a0this type of problem, we have developed a novel paradigm called distribution\u00a0temporal logic (DTL) which can be used to describe such tasks. DTL is a\u00a0temporal logic defined over sample paths of random processes (e.g. the unknown\u00a0underlying state of the environment) and functions of estimate distributions\u00a0(e.g. measures of uncertainty or expected costs). DTL can be used to express\u00a0such missions as “Explore the environment until the entropy of the estimate of\u00a0the survivor locations is below 5 bits. If a survivor is found in an unsafe location with probability 0.8, guide it to a safe region.” DTL is defined in\u00a0full here<\/a>. Our current research focuses on developing algorithms to find\u00a0control policies that maximize the probability of satisfying a given DTL\u00a0specification.<\/p>\n <\/p>\n","protected":false},"excerpt":{"rendered":" Active sensing is the problem of \u00a0coordinating a robot or team of robot’s motion and sensor alignment such that they can estimate some spacially distributed feature of the environment. \u00a0 An example of such a problem is a team of aerial robots with onboard cameras being used to monitor traffic conditions in an urban environment. […]<\/p>\n","protected":false},"author":7535,"featured_media":0,"parent":44,"menu_order":6,"comment_status":"closed","ping_status":"closed","template":"","meta":[],"_links":{"self":[{"href":"https:\/\/sites.bu.edu\/msl\/wp-json\/wp\/v2\/pages\/113"}],"collection":[{"href":"https:\/\/sites.bu.edu\/msl\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/sites.bu.edu\/msl\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/sites.bu.edu\/msl\/wp-json\/wp\/v2\/users\/7535"}],"replies":[{"embeddable":true,"href":"https:\/\/sites.bu.edu\/msl\/wp-json\/wp\/v2\/comments?post=113"}],"version-history":[{"count":11,"href":"https:\/\/sites.bu.edu\/msl\/wp-json\/wp\/v2\/pages\/113\/revisions"}],"predecessor-version":[{"id":355,"href":"https:\/\/sites.bu.edu\/msl\/wp-json\/wp\/v2\/pages\/113\/revisions\/355"}],"up":[{"embeddable":true,"href":"https:\/\/sites.bu.edu\/msl\/wp-json\/wp\/v2\/pages\/44"}],"wp:attachment":[{"href":"https:\/\/sites.bu.edu\/msl\/wp-json\/wp\/v2\/media?parent=113"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}