This bachelor thesis is about the design of the Safety and Sensing module, which is part of the Environment Observation Module for Deci Zebro robots. This Environment Observation Module is developed at the faculty of Electrical Engineering, Mathematics and Computer Science (EEMCS) of Delft University of Technology. Zebro robots were not able to walk around without falling off things or without collide against obstacles. Also Zebro was not able to collect data of its surrounding or of its own position and orientation. With the Safety and Sensing Module Zebro will be able to collect data of its surroundings such as light intensity, humidity and temperature and is equipped with lift and fall detection and over-current and internal overheat protection. Also the safety of Zebro and its ability to walk around safely are improved with this module. Because Deci Zebro is not able to walk at this time, the module is not tested under real circumstances. In simulated circumstances the module works as intended.

In this thesis we will walk you through the design choices we made in designing a battery management system (BMS) for the Zebro robot. This specific BMS is designed as a subcomponent of an autonomous charging system. The main goal of this BMS is to safely charge and discharge the systems battery, while providing the system with relevant data on the battery status. The BMS consists of multiple subsystems, the most important being a charger, a cell balancer, a power management system and a main control unit tying it all togheter.

The Zebro is a six legged autonomous robot, designed to be deployed in swarms. Charging the battery is still done manually. To further increase the autonomous abilities of the Zebro, a wireless charging station and module is designed. The charging station is fully modular and interconnected by charging pads. This opens the possibility of charging multiple Zebro's simultaneously. The charging module consists of a receiver and Battery Management System inside the Zebro. In this thesis, both the transmitter and receiver side of the Wireless Power Transfer (WPT) system is designed. The transmitter and receiver must be perfectly aligned for high efficiency in a wireless charging setup. The positioning of the Zebro is not highly accurate, therefore the design of an inductive WPT system using SS-topology compensation is presented. Also, possible solutions improving the efficiency of the power transfer and the WPT system is proposed in this thesis. These improvements include coil optimization, object detection, circuit damage protection and implementation into the overall system.

The ZEBRO is a fully autonomous six-legged robot designed for swarm behaviour and is the size of an A4 sheet of paper. However, there does not exist an charging station for the ZEBRO yet. At this time, the battery still has to be replaced manually, which defeats the purpose of an autonomous system. An autonomous charging station has to be designed to complete the system. The charging station can be split into 3 parts: the autonomous charging station, the wireless power transfer, and the battery management system (located in the ZEBRO). In this particular thesis, the autonomous charging station will be discussed. Since all components in the ZEBRO are built to be modular, the design of the autonomous charging station is chosen to fit this principle. The charging station will consist of one central unit, and several connectable (modular) pads. Each pad will be able to hold and charge a single ZEBRO in 30 minutes. The charging station contains a communication module to communicate its location to ZEBRO's in need of charging. To calculate the locations of the connected pads, a grid will be created by serial communication between the charging station and the connected pads through Raspberry Pi's. (The pads do not have a communication module and do not have their location defined) To power the wireless power transfer unit, a power supply of 48V and 500W will be implemented in the charging station. Each pad will consume about 100W, so a maximum of 4 pads can be connected. (Due to power loss). If more pads need to be connected, an additional power supply can be put in parallel. All separate components have been simulated or tested and do work as predicted. The integration of all components of the charging station was successful as well. At this time, the complete system with the wireless power transfer and battery management system has not yet been put together and so it is unknown if it works as a whole.

A Camera Submodule for the Environment Observation Module for a Zebro robot was designed. It is able to collect and process data about obstacles on a Raspberry Pi and send this to the main processor of the Environment Observation module using UART. The two main functions that were implemented are a laser rangefinder based on triangulation and a motion detector based on optical flow. The laser rangefinder calculates the distance to an obstacle by finding its own laser point in a frame. Testing showed that it is especially accurate at short distances (0-4 cm), and the range can differ between 0.30 to 6m due to environmental light and obstacle color. The motion detector uses dense optical flow data that is provided by the h264 encoder of the Pi Camera. This data is filtered and moving obstacles are grouped together. In the tests, a moving obstacle was detected every time, but 29% also showed false positives due to large egomotion.

In this thesis, we describe the design process of a distance sensing system for the Deci Zebro swarm robots. We use a technique that transmits a radio frequency message and a ultrasonic pulse concurrently. Due to the difference in propagation speed of both signals, the distance could be measured using time difference of arrival (TDOA). A cone shaped antenna is designed to create a 360 ultrasonic pulse coverage. At the end of this thesis we present a prototype with a range of 7 m. We find a linear relation between the TDOA and the actual distance between the modules. We thus conclude that our prototype is suitable for range measurements on roving swarm robots.