Tobias Mann

Automating planning and scheduling of shuttle payload operations

Abstract This paper describes the DATA-CHASER Automated Planner/Scheduler (DCAPS) system for automated generation and repair of command sequences for the DATA-CHASER shuttle payload. DCAPS uses general Artificial Intelligence (AI) heuristic search techniques, including an iterative repair framework in which the system iteratively resolves conflicts with the state, resource, and temporal constraints of the payload activities. DCAPS was used in the operations of the shuttle payload for the STS-85 shuttle flight in August 1997 and enabled a 80% reduction in mission operations effort and a 40% increase in science return.

Strategies for autonomous rovers at Mars

The science return from future robotic exploration of the Martian surface can be enhanced by performing routine processing using onboard computers. This can be accomplished by using software that recognizes scientifically relevant surface features from imaging and other data and prioritizes the data for return transmission to Earth. Two algorithms have been designed and evaluated with field data to identify the properties of the environment that can be reliably detected with onboard imaging and multispectral observation. One algorithm identifies variations in surface textures in images and successfully distinguishes between rocks and soil and between differences in grain size in a rock of a single composition. A second algorithm utilizes a neural net to recognize selected carbonate minerals from spectral reflectance data and successfully identifies carbonates from a set of spectra collected in the field. These types of algorithms will contribute to the efficiency of a landed instrument suite given the limited resources of time, data storage, and available communications opportunities.

Texture Analysis for Mars Rover Images

The problem addressed in this paper is that of clustering image pixels into regions of homogenous geological texture. Future rovers on Mars will need to be able to intelligently select data collection targets. One goal of intelligent data selection for maximizing scientific return is to sample all distinct types of rocks that may be encountered. Different rock types often have a characteristic visual texture, thus visual texture is rich source of information for separating rocks into different types. Recent work on using texture to segment images has been very successful on images with homogenous textures such as mosaics of Brodatz textures and some natural scenes. The geologic history of a rock leads to irregular shapes and surface textures. As a result, the textures in our images are not as homogeneous as those in Brodatz mosaics. Our approach is to extract textural information by applying a bank of Gabor filters to the image. The resulting texture vectors are then clustered. Banks of filters constrain the relationships of the filter parameters both within a single filter and between filters. Often researchers have used parameter values that are thought to correspond to the human visual system, however the effects of adjusting these parameters have not been thoroughly studied. We systematically explore tradeoffs in the parameter space of the filter bank and quantify the effects of the takeoffs on the quality of the resulting clusters.

Interactive, repair-based planning and scheduling for Shuttle payload operations

This paper describes the DATA-CHASER Automated Planner/Scheduler (DCAPS) system for automatically generating low-level command sequences from high-level user goals. DCAPS uses Artificial Intelligence (AI)-based search techniques and an iterative repair framework in which the system selectively resolves conflicts with the resource and temporal constraints of the DATA-CHASER Shuttle payload activities.

Morphogenesis in plants: modeling the shoot apical meristem, and possible applications

A key determinant of overall morphogenesis in flowering plants such as Arabidopsis thaliana is the shoot apical meristem (growing tip of a shoot). Gene regulation networks can be used to model this system. We exhibit a very preliminary two-dimensional model including gene regulation and intercellular signaling, but omitting cell division and dynamical geometry. The model can be trained to have three stable regions of gene expression corresponding to the central zone, peripheral zone, and rib meristem. We also discuss a space-engineering motivation for studying and controlling the morphogenesis of plants using such computational models.