Micro-helicopters rely on global communication to receive orders and commands. The same mean is used to convey status reports and possibly images to a remote center. In addition, local communication supports the exchange of messages between the helicopters for coordination purposes. One of the information that is critical to them is an estimate of the distances between them. Even if a GPS receiver may be installed on each helicopter to provide global positioning, an access to some relative positions, in particular along the vertical axis, is very important. GPS receivers have a high power consumption and relatively low accuracy in the z axis. The objective of the work package is three fold:
WP3.1: Development of local communication system including distance estimation
Local communication will rely on the technologies developed under the name of “wireless sensor networks”. There will be no hardware development within the project. Rather, we will reuse one the modules available at CSEM and possibly adapt it. These modules offer short-range communications (<2km in open space) with limited power consumption (80mW maximum) in the 2.4GHz and 868/900 MHz bands. Their volumes do not exceed a few cubic centimeters. Raw bit rates up to 250 kbit/s are available. With the configuration of the flying vehicles, the Medium Access Control (MAC) is the most important protocol layer. Routing may also be needed but is less of a challenge. The work will thus concentrate on the development of a medium access protocol (MAC) specially designed for fast interaction and low power consumption. To avoid any central point of failure and because fleet coordination is a distributed problem per se, the MAC will not rely on a central coordinator. Similarly, the dynamic nature of the problem precludes using MACs that are based on global synchronization and TDMA like communication scheduling. The work will start from the WiseMAC protocol developed at CSEM. WiseMAC is a medium access control protocol designed for wireless sensor networks. This protocol is based on non-persistent CSMA and uses the preamble sampling technique to minimize the power consumed when listening to an idle medium. The novelty in this protocol consists in exploiting the knowledge of the sampling schedule of one’s direct neighbors to use a wake-up preamble of minimized size. This scheme allows not only to reduce the transmit and the receive power consumption, but also brings a drastic reduction of the energy wasted due to overhearing. WiseMAC requires no set-up signaling, no network-wide synchronization and is adaptive to the traffic load. It presents an ultralow power consumption in low traffic conditions and a high energy efficiency in medium traffic conditions. WiseMAC protocol outperforms the majority of WSN MAC protocols under low to medium loads. The objective of the task is to enhance WiseMAC towards higher loads while keeping high responsiveness and low power consumption. One of the possible approaches is to design a protocol that can dynamically adapt its operating point to privilege responsiveness or latency when needed and return to a low power consumption mode when mobility is reduced. As this might not be sufficient, we will also explore new protocols that allow rapid and fugitive interactions under low power consumption.
In conjunction with the investigation on the MAC protocol, we will port the relative positioning techniques based on RSSI (Receive Signal Strength). The challenge is to obtain fast measurements under low consumption and relatively fast moving objects. The idea is to combine the MAC protocol with the relative positioning to avoid duplicating information and thus reduce overhead to a minimum.
WP3.2: Integration of highly integrated GSM communication unit
Global communication will use the GSM network. A number of communication modules are available on the market. We will use one of these modules. Because of the strong weight constraints, the module should be selected carefully to find the best compromise between size and power consumption (as power consumption translates into battery weight). The module peak current is also a criteria (peak currents on up to a few amps are not rare). Once selected, the module will be integrated with the helicopter controller and the necessary driver written.
WP3.3: Integration of communication system on micro-helicopter
This last task (in sequence) deals with the integration from the hardware and software viewpoint of the GSM module and local communication module to the micro-helicopter. It will include unit testing of the combination.